Universiteit Van Amsterdam
The effect of market capitalization on US equity REITs
performance
Brian Moulton
11391529
Bachelor of Economics and Business Economics Thesis
Thesis Supervisor:
Dr. M.I. Dröes
University of Amsterdam
Faculty of Economics and Business
Specialization : Finance
Abstract
REITs are the second largest source of investment into real estate at 24% and this number continues to grow, thus understanding the determinants of REIT performance is becoming increasingly important. This thesis goes on to evaluate how market capitalization effects EREIT performance. The results show a significant positive relationship between market capitalization and EREIT performance, controlled for the real interest rate, S&P 500, real GDP growth and population, which is tangent with previous research between market capitalization and firm performance. This study benefits existing literature by providing current results, as well as narrowing down previous research between market capitalization and firm performance by making it generalizable to EREITs specifically. These results allow investors to make more informed decisions about investing in EREITs and thus have a more balanced, diversified portfolio.
Statement of originality
This document is written by, B.M. Moulton, who declares to take full responsibility for the
contents of this document. I declare that the text and the work presented in this
document is 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 of the University of Amsterdam is responsible
solely for the supervision of completion of the work, not for the contents.
Table of contents
1. Introduction 4
2. Literature review 6
2.1 Brief history of REITs 6
2.2 Characteristics of REITs 7
2.21 Interest Rates and REITs 7
2.22 Systematic Risk and REITs 8
2.23 GDP, Population and REITs 9
2.3 Market Capitalization 9 2.4 Hypotheses 10 3. Data 11 3.1 Dependent Variable 11 3.2 Independent Variable 11 3.3 Control Variables 13 4. Methodology 15 4.1 OLS Regression 15 5. Results 16 6. Conclusion 19 7. References 20
1. Introduction
The ‘american dream’ is what people refer to as the equal sets of opportunities presented within the US, allowing the most hard working people to rise to the top of the economic ladder. This can be done in any amount of ways but the leading category of millionaires in the world is those in real estate, with 90% of today's millionaires having been created through investment in real estate. One such way of investing in real estate is through a REIT, a publicly traded company that owns income producing real estate, of which it pays out more than 95% as dividends to shareholders. REITs are the second largest source of investment into real estate, only behind private equity. To put into perspective the sheer size of REITs as an investment vehicle, as of 2018, the market capitalization of New York Stock Exchange listed REITs summed to $955.8 billion, spread over 191 publicly traded REITs. In 2017 alone, publicly traded REITs paid out in excess of $53 billion as dividends (Gaille, 2018) . In addition to the sheer size of REITs, Bhuyan et al. (2015) found REITs, especially EREITs to be a helpful tool for portfolio diversification, they provide a method for all investors to invest into real estate which bypasses the previously high barriers to entry. Due to the reasons mentioned above it is crucial to understand how the REIT market works. One way of doing this is to look at what factors affect performance of REITs, including but not limited to the market capitalization.
Previous studies have looked into multiple variables that influence REIT performance. Beginning with Chen and Tzang (1988), who looked at the effect of long term real interest rates on REIT returns, they found there to be a significant effect but the differences over time were so great that further research was required to evaluate these time differences. Research continued and it was found that a stock market index was better at predicting REIT returns (Swanson et al., 2002). Additional studies went on to find that both GDP (Kolz and Kodongo, 2017) and Population (Mulder, 2006) had significant effects on REIT returns. Other studies went on to look at the effect of market capitalization on firms performance. Subeniotis et al. (2011) found there to be a large, statistically significant effect of market capitalization on firms performance. However no research was done that looked at market capitalization and REIT performance. This provided a knowledge gap, which this Thesis looks to solve by answering the research question:
‘To what extent does market capitalization affect the performance of US residential EREITs’
Multiple databases were used to collect the necessary data. The span of the study was 10 years, from 2010 to 2019 with information on 100 separate REITs each with yearly data points, amassing to 1000 data points. The dependent variable, ROE, is used to measure a REITs performance. The independent variable is market capitalization, both ROE and market capitalization are obtained from Factset for 100 REITs. The control variables are the real interest rate, the S&P 500, real GDP growth and population. Data for these variables was obtained from the ‘Federal Reserves System’, the ‘US Bureau of Economic Analysis’ and Factset.
Using this data two OLS regressions were run to conduct the effect of market capitalization on EREIT performance. The dependent variable, ROE, was regressed only on market capitalization for regression 1, and in regression 2, ROE was regressed on market capitalization and all other control variables mentioned above. In addition to this the study includes a descriptive statistics table, a correlation matrix and more graphs to help analyse the data and draw a conclusion.
The results show a positive effect of market capitalization on EREIT performance in both regressions, for regression 1 the significance level is 0.1% and for regression 2 the significance level is 1%. This is in line with previous research that found market capitalization to have a positive effect on firm performance, this research has extended that research to specifically EREITs.
This thesis contributes to existing research by looking at a specific category of firms, namely EREITs. It shows that the existing research holds, and with small differences can be extended to generalise for EREITs too. In addition to this, the research is current, with no data having been collected more than 10 years ago, it provides a clear view of how EREITs are acting today.
To answer the research question, the thesis layout is as follows. First there will be a literature review on REITs, the factors affecting their performance and market capitalizations effect on firm performance. Next, there is an overview of the data, followed by the methodology where the specifics of the regression will be addressed. This will be followed by a presentation of the results and finally a conclusion which will answer the research question.
2. Literature review
In this section, there will be an overview of current knowledge of REITs and market capitalization. Beginning with describing REITs, how they work and the factors affecting REIT returns. Followed by the effects of market capitalization on businesses and concluding with research that links market capitalization and REIT returns.
2.1 Brief history of REITs
Real estate investment trusts, referred to as REITs, are pooled real estate funds giving individuals access to different types of investments within the real estate industry. Although there are multiple forms of REITs, the most popular forms are equity REITs (EREITs) and mortgage REITs (MREITs), differing in the assets they own. EREITs focus on holding real estate properties and pay out income from these properties as dividends to shareholders, MREITs on the other hand hold loans related to real estate such as construction loans and mortgages, also paying out income to shareholders in the form of dividends. Dividend yield accounts for the majority of a REITs return as REITs that payout 95% or more of their income as dividends are exempt from corporate income tax. This means that shareholders can avoid double taxation on their earnings, but this is not the case for the REIT, any income that has not been paid out as dividends can be taxed at the corporate income tax level. REITs do not generally experience any problems here as this taxation will be on max 10% of their income in the most extreme of circumstances (Chen and Tzang, 1988).
Semer (2009), argues that REITs were put in place to fulfill two financial goals. Firstly, they allow small investors to diversify their portfolio by investing in a new asset class. Meaning all investors, not just institutional investors now have access to diversification through real estate investment and diversification within real estate via REITs. Secondly this inflow of pooled capital means a new source of financing for real estate developers that was previously not accessible.
Further studies examined differences in the diversification effects between EREITs and MREITs and whether the risk mitigation was worth holding these REITs within a portfolio. Bhuyan et al. (2015), found EREITs to be a helpful tool for portfolio diversification but found MREITs to be unnecessarily risky compared to EREITs.
The tax reform act of 1993 by the Clinton administration brought about many changes within the REIT industry. Pre 1993, a domestic pension plan was treated as an individual plan regardless of the number of participants. In addition to this, to qualify as a REIT, firms had to adhere to the 5/50 rule which stated that in the last half of its taxable year 5 or less people could not own more than 50 percent of the value of the firm. This meant that due to their large nature, pension plans, regardless of the number of participants, could not hold REITs so as to break this 5/50 rule. Post 1993, when the tax reform act was put into place (January 1st 1994) ‘individual pension plans’ were now looked into, to evaluate how many participants they truly had. These structural changes gave rise to institutional investment within REITs, forever changing the way that REITs would perform (Glascock et al., 2000).
2.2 Characteristics of REITs
2.21 Interest Rates and REITs
Multiple scholars have looked into the relationship between interest rates and REIT returns. Chen and Tzang (1988) were some of the first to delve into this relationship. Their study, along with the results focused on differentiating between REITs' relationship with long term interest rates versus short term interest rates and how these effects differ over time periods. Their study spanned over 12 years, from 1973-1985 and it was found that during the period 1973-1979 both equity and mortgage REITs had a relationship with only long term interest rates while in the 1980-1985 period both REIT categories had a relationship with long term and short term interest rates. In conclusion the research showed that the interest rate, especially the long term interest rates had a statistically significant effect on REIT returns but that more research had to be done into the difference in effects over time periods.
Swanson et al. (2002) looked to expand on previous research by looking at how REIT returns correlated with interest rates and a stock market benchmark. Their findings showed that interest rates did have an effect on REITs but the level of affect varied over time, they also found that the value-weighted stock market index did a better job at predicting REIT returns. This conclusion supports previous work and expanded on it by showing that REIT returns and a stock market index were highly correlated.
Most recent research by Giliberto & Shulman (2017) found that interest rate sensitivity to EREITs varies over time and that there is no long run predictive rule to calculate how returns change due to interest rates.
2.22 Systematic Risk and REITs
Economic theory states that the beta of a stock within the capital asset pricing model shows the correlation between the stock's return and the market return, a firm's systematic risk can be measured using its beta. Knowing the beta of a firm can help with diversification of a portfolio amongst other things.
Patel and Olson (1984) noticed that previous research only managed to explain beta averages within industries and not individual firm betas. Their research aimed to be sector specific to real estate investment trusts and thus allow them to explain systematic risk of specific firms within the REIT sector via financial determinants. The results showed that financial determinants, such as advisor fee and financial leverage were significantly positively correlated with systematic risk within REITs. This development of systematic risk within REITs led more scholars to look at the correlation between REIT returns and the market returns
Delcoure and Dickens (2004) added to past work by differentiating systematic risk between REOCs (Real Estate Operating Companies) and REITs, the main difference between these two investment types being that REOCs held a large percentage of their properties in hotel assets.. Although at first glance the data seemed to show that REOCs had consistently lower systematic risk then REITs, upon closer inspection it was found that if the sample of REITs was controlled by term structure and investing in the same percentage of hotels as an average REOC the results showed that REOCs had significantly higher systematic risk than REITs. Further results showed that REIT systematic risk had a negative correlation with ratio of funds from operations to total assets and that REITs with greater industrial property holdings are exposed to less systematic risk.
Further research into the stability of betas for individual EREITs showed that between 1972-2002 betas remained steady with a significant decline in 2002, however the reason for this decline is not definitive, it could be sampling error or an actual significant decline (Chiang et al. 2005).
Overall, under the null of the three factor model, EREIT returns show long term stability in relation to market risk.
2.23 GDP, Population and REITs
Two more factors that are correlated with REIT returns are GDP and population. Kola and Kodongo (2017) found that REITs tend to follow the business cycle, stipulating that they benefit from a growing economy and are negatively affected by a contracting one. They use GDP as a base to follow the business cycle and found that REIT returns are positively correlated with GDP. This makes logical sense as when an economy is booming there are more investment opportunities within real estate, cash rich individuals and plenty of new rental agreements.
Mulder (2006) looked at housing and population. It was found that the relationship is two sided, with the one side being changes in population influencing housing demand and the other side being changes in the supply of housing causing a change in the environment for migration and new household formulations. This research can be stipulated to say that population has an effect on REITs, the direction of this affect however is unknown.
2.3 Market Capitalization
The stock market is a financial market used by firms to raise capital, its main purpose being to join lenders and borrowers for the financing of industries who lack current capital. A specific firms ‘market capitalization’ can be taken as the total value of this company that is being traded on the market. This can be calculated by multiplying shares outstanding by current share price. Consisting of the primary market, where firms can issue new securities and the secondary market, where already existing securities are traded. Research by Odularu (2009) looking at market capitalization in the nigerian confectionery industry found that market capitalization and firm performance have a positive relationship. Both in terms of market capitalization as a whole as well as specifically the share price, two different parts of one whole.
Further research was conducted by Subeniotis et al. (2011) and this time, the data used was a panel of 12 european countries between 2000 and 2005 that formed an index. Their findings supported past literature showing that market capitalization has a high and significant positive correlation with firms' performance.
Before looking at how market capitalization affects firms performance it is important to understand what internal firm factors affect market capitalization. Garcia and Liu (1999) were the first to look at this relationship, using a dataset of 15 developing and industrial countries between 1980 and 1995. It was found that financial intermediary development, stock market liquidity, real income levels and the savings rate were all statistically significant indicators of market capitalization.
2.4 Hypotheses
After evaluating the literature it can be seen that market capitalization has a high, statistically significant positive relationship with firm performance in many different markets. In addition to this, there are multiple known factors such as real income levels that affect a firm's level of market capitalization.
Although there has been plenty of research looking at the effects of market capitalization on firm performance there is not any current research since 2011. In addition, the research on the effects of market capitalization has not been specified to EREITs. This shows a knowledge gap and means that a variable that has been proven to have an impact on firm performance may also have an impact on REIT performance, knowing whether this is the case or not will help investors make more informative decisions on which EREITs to hold depending on the outcome. This thesis will look directly at the impact of market capitalization on EREITs performance between 2010-2019. For these reasons, this thesis will be beneficial to existing literature.
Thus my research will address ‘To what extent does market capitalization affect the performance of US residential EREITs’. I expect to see that market capitalization has a positive and significant effect on ROE of EREITs, even more so than the already known correlation between market capitalization and firm performance in other markets. I expect this to be the case because the large market cap of an EREIT gives it access to many additional investment opportunities that it would not have access to given a smaller market capitalization.
3. Data
In this next section all the data used throughout the thesis will be explained. Starting with the dependent variable, then the independent variable and finally the control variables. It will go into detail on how the data was gathered, descriptive statistics and why this data has been included.
3.1 Dependent Variable
To gage the performance of EREITs, their return on equity will be used. The data is obtained from Factset for 100 US EREITS each with 10 data points, totalling for 1000 points over a span of 10 years, from December 2010 until December 2019. For consistency reasons, all data points were obtained at the end of the fiscal year. The variable return on equity will be referred to as ROE from here on out and will always be in percentage terms. As seen in table 1 the mean is 6.169% with a standard deviation of 10.4%. From graph 1, which was computed by finding the average of all 100 firm's ROE in separate years, it can be seen that the average ROE has been steadily rising between 2010 until 2016 where growth slowed until it’s periodic high of 8.8% in 2019.
Graph 1 - Average ROE
3.2 Independent Variable
data was gathered over the same period and the data points are also at the end of each fiscal year. Market cap is computed by finding the shares outstanding multiplied by the stock price and is measured in millions of USD. As seen in table 1, the mean for market cap is 5312 with a standard deviation of 8535. From graph 2, which was computed by finding the average of the 100 firms market cap in separate years, the average market cap steadily rose between 2010 and 2017, dropped in 2018 and somewhat recovered in 2019. Throughout this thesis it will be evaluated whether the variable market cap has a significant effect on EREIT returns.
Graph 2 - Average Market Cap
Table 1 - Descriptive Statistics
N Mean Std. Dev. Min Max
ROE 1000 6.169 10.393 -57.630 126.900
Market Capitalization 1000 5311.835 8534.802 5.790 101784.980
Real Interest Rate 1000 0.846 0.377 0.220 1.730
S&P 500 1000 2054.270 609.646 1257.60 3230.78
Real GDP Growth 1000 2.280 0.458 1.600 2.900
Population 1000 319.578 6.123 309.774 328.527
3.3 Control Variables
Four control variables are included in the data, specifically the real interest rate, S&P 500, real GDP growth and population.
Firstly, the real interest rate is the 20 year inflation adjusted yield of US treasury bonds and the data was collected from the ‘Federal Reserves System’. The reason for using the inflation adjusted yield is to find the real level of interest rate as this is more accurate when studying its effect on ROE. The 20 year yield was taken because this can be considered the ‘long term’ real interest rate, and as EREITs are a long term investment this is a suitable time period for real interest rates. Over the 10 year sample period real interest rates steadily decreased from 2010 to 2019. This is likely because post financial crisis of 2007/8 the US government was looking to stimulate economic activity by driving down interest rates. This control variable is included because it has been found in previous research that the interest rate has a negative relation to EREIT returns (ROE). This can also be seen in table 2 where the effect of the real interest rate on ROE is indeed negative, in addition to this the coefficient is -0.192, showing a weak correlation.
Secondly, the S&P 500 Price Index has been used as a control variable. The data was collected from Factset and is an index that tracks the 500 largest companies in the US by market capitalization, it is widely considered to be the best representation of how the US stock market is performing. Throughout the 10 year sample period the S&P 500 has shown an upward trend from $1258 in 2010 up to $3231 in 2019, this is as expected as there have been no recessions between this period and growth in the US was moderate and steady. The S&P 500 is used as a control variable because multiple studies have found there to be a significant positive relation between how the S&P 500 performs and how EREITs perform. This can also be seen in table 2 as there is a positive correlation of 0.206.
The third control variable used is real GDP growth for the US, the data was collected from the US Bureau of Economic Analysis (BEA) and it measures the change in the value of goods and services produced in the US. Within the data it can be seen that the real GDP growth in the US over the period 2010 to 2019 was constantly positive, fluctuating between 1.6 (2011) and 2.9 (2015), standard levels for the US (not including during periods of economic stress). This has been included as a control variable because GDP growth is a good indicator of how the economy as a whole is performing, high levels of GDP growth transfer to increasing economic performance, and thus higher EREIT performance (ROE) and vica versa. As seen in table 2 there is a positive correlation between real GDP growth and ROE of 0.0569, showing a weak yet positive correlation between the two variables.
The fourth and final control variable included in the data set is population, it too is collected from the US Bureau of Economic Analysis and is measured as the total population of the US at the end of fiscal year, measured in millions. As expected the US has
from circa 310 million at the very beginning to a maximum of 328.5 million. Population has been used as a control variable because as the population increases, so does the demand for real estate, be it offices, commercial or apartments. This increase in demand positively affects the ROE for EREITs and thus we expect a positive correlation between the two variables. This is also shown when looking at table 2 which shows a positive coefficient between the two variables of 0.219, both moderate and positive.
Table 2 - Correlation Matrix
ROE Market Capitalizati on Real Interest Rate S&P 500 Real GDP Growth Populatio n ROE 1 Market Capitalization 0.135 1
Real Interest Rate -0.192 -0.183 1
S&P 500 0.206 0.199 -0.879 1
Real GDP Growth 0.0569 0.0527 -0.0938 0.245 1
Population 0.219 0.198 -0.897 0.962 0.271 1
Table 2 shows a correlation matrix between all the variables in my model. A score of 1 represents a perfect correlation between the two variables and a value of 0 represents no correlation between the two variables. For consistency reasons all my data has been collected over the same time period, of 2010-2019 and using this data it will be looked at what affect market capitalization has on the ROE of EREITs.
4. Methodology
4.1 OLS Regression
Two OLS regressions will be used to answer the research question, ‘To what extent does market capitalization affect the performance of US residential EREITs’. This model works by finding a ‘line of best fit’ which minimizes the sum of squares for the dependent variable between the observed values and the values generated using the model. It will be
used as it allows us to see whether there is a significant relationship between the dependent (ROE) and independent variable (Market Capitalization). The F-test will be used to evaluate the model as a whole and the p-values will be used to evaluate the individual variables.
The first linear regression is listed below and will consist of solely two variables, being ROE and Market Capitalization.
β β MC ROEt = 0 + 1 t
The second regression is a multiple regression model as listed below and will consist of 6 variables namely, ROE, Market Capitalization, Real Interest Rate, S&P 500, Real GDP and Population.
β β MC β RIR β SP β GDP β P ROEt = 0 + 1 t + 2 t + 3 t + 4 t + 5 t
The two models will be used to answer the research question, which has been represented mathematically below. H0represents the first hypotheses, which states that the coefficient of market capitalization is equal to 0, if this hypothesis holds this means that market capitalization has no significant effect on ROE for EREITs. If however H0 does not hold then by reasoning, H1 must hold. This would mean that the coefficient of market capitalization is not equal to 0, and thus market capitalization has a significant effect on ROE.
β 0 H = H0 : 1 = 1: β1 / 0
Table 3 – List of Variables ROEt
Return on Equity at time t MCt
Market Capitalization at time t RIRt
Real Interest Rate at time t SPt
S&P 500 Price Index at time t GDPt
Real Growth GDP at time t Pt
5. Results
Throughout this section the statistical results of the regressions will be interpreted by looking at the different outputs from the two OLS regressions (Table 4). First we will evaluate regression 1, followed by regression 2 The coefficients, their signs and their statistical significance will be tested, followed by evaluating the models as a whole. Using this information it will be possible to move onto the conclusion for this Thesis.
Looking at regression 1, the constant of the model has a coefficient of 5.299, which is statistically significant at 0.1% meaning that theoretically if the company had 0 market capitalization it would expect to have a ROE of 5.299. As you can see this is not possible and thus isn’t of particularly much use to us. The independent variable market capitalization has a positive coefficient of 0.000164. This means that for every increase of market capitalization by 1 million USD, the model predicts a 0.000164 increase in ROE. At first glance this doesn’t seem particularly large but from table 1 it can be seen that the mean market capitalization for these EREITS is 5311.835, so an increase in market capitalization of 1000, which is realistic over some years, would cause a predicted increase in ROE of 0.164%. The p-value for market capitalization is smaller than 0.001 meaning that the effect is significant at 0.1% significance level. This means that for a significance level of 0.1% H0 can be rejected meaning β1 is different from 0 and thus market capitalization has an effect on ROE.
The adjusted R-Squared for regression 1 is 0.0171, this means that the model only predicts 1.71% of the variance of ROE. Looking at the F-statistic, F(1,998) = 18.411 means that for a significance level of 0.1%, the null hypothesis can be rejected (we can use F-test because there is only one variable in this regression) and thus the market capitalization has an effect on ROE.
Looking at regression 2, the constant has a coefficient of -156.230 which is statistically significant at 5%. This means that if hypothetically all other variables had a value of 0 (which is impossible given our data set), the ROE for EREITs would be predicted at -156.230. This constant is not of use to us but because it has such a large negative value it means that our other variables will have a large effect on ROE.
In regression 2, the coefficient for market capitalization is 0.000117, which is lower than that of regression 1, and expected, as it is likely that market capitalisation had a correlation with the other variables and adding these decreased the coefficient as they now predict the ROE more accurately. This coefficient means that for an increase of 1 million
USD to the market capitalization of an EREIT, this model predicts an increase of 0.000117% increase in the ROE. The p value is smaller than 0.01 meaning that the effect is significant at 1%. This means that at 1% significance the null hypothesis is rejected and market capitalization can be said to have a positive effect on ROE for EREITs. These results show the same as the literature on the relationship between market capitalization and firms performance, both positive and significant.
The real interest rate has a coefficient of 0.802 which means that for an increase of the real interest rate by 1%, the model predicts an increase of ROE of 0.802%. The p-value for real interest rate is larger than 0.05, meaning that for a significance level of 5% we can not say that the results are statistically significant, this may be due to the fact that the sample size was too small or more likely because there is too much variability within the data to predict it. This finding is not in line with previous research of Chen and Tzang (1988) , which found the long term real interest rate to have a statistically significant effect on firm performance (in this model ROE), this may be because ROE is not a suitable measure of firm performance. The sign of the coefficient supports previous findings, where research found the relation to have opposite effects depending on the time period, supposedly from our findings the period 2010 to 2019 has a positive relation between real long term interest rates and ROE.
The next coefficient to look at is that of the S&P 500, the coefficient is -0.00131 and the p-value is greater than 0.05, thus for a significance level of 5% the coefficient is not statistically significant. This means that for a 1 unit increase in the price index of S&P 500, the model predicts a decrease in ROE for EREITs of -0.00131%. This can also be looked at as EREITs having a negative beta within the CAPM model. However these findings can not be said to be statistically significant due to the high p-value.
Real GDP growth has a coefficient of -0.193, which means this model predicts a decrease of ROE for EREITS of -0.193% for a 1% increase in real GDP growth. This does not support previous research by Kola and Kodongo (2017) ,who found that GDP growth has a positive effect on firm performance. This may be because ROE is not a suitable measure of firm performance. But still seems bizarre considering EREITs have been found to be procyclical in the past. The p-value of real GDP growth is larger than 0.05, meaning that for a significance level of 5% the coefficient is not significant.
The final coefficient for population is 0.514, meaning that the model predicts an increase of ROE for EREITS of 0.514% for an increase of the population by 1 million people. The p-value of the coefficient is less than 0.05, meaning that the coefficient for population is
has a significant effect on REIT performance, and builds on it by finding that it is a positive effect and also significant for specifically EREITs.
The adjusted R-Squared for regression 2 is 0.0527, meaning that the model predicts 5.27% of the variance of ROE for EREITs. This value is considerably higher than that of regression 1 as expected due to the higher predictive power of the additional variables added to the regression. The F-statistic, F(5,994)=12.12 means that the model is significant for a significance level as low as 0.1%.
Table 4 – OLS Regression Results
Dependent Variable (Regression 1) ROE (Regression 2) ROE Constant 5.299*** (0.384) -156.230* (66.430) Market Capitalization
Real Interest Rate S&P 500 Real GDP Growth Population #of Observations R-Square Adjusted R-Square F-Statistic 0.000164*** (0.0000382) 1000 0.0181 0.0171 18.411 0.000117** (0.000383) 0.802 (2.069) -0.00131 (0.00195) -0.193 (0.775) 0.514* (0.217) 1000 0.0575 0.0527 12.12 Note: robust standard errors in parentheses; *p<0.05, **p<0.01, ***p<0.001
6. Conclusion
This thesis has evaluated the influence of market capitalization on the return on equity for equity real estate investment trusts by analysing the outputs of two different OLS regression models. The research used data from a 10 year time period, 2010 to 2019. It has benefited existing research by expanding from the generalized relations between market capitalization and ROE to a more focused approach specifically for EREITs, in addition to this it has used current data, further adding to the value.
The results show that market capitalization has a significant positive effect on the performance of EREITs, which was measured by ROE. Within the bounds of both regressions this was found to be the case. These results mean that the performance of EREITs can be predicted more effectively if including market capitalization as a predictor, this provides significant value when choosing which EREITs to invest in whether at an institutional level or individually.
A limitation for this research is the frequency of the data points, being only yearly, changing this frequency to monthly or even daily would measure the effect of market capitalization on EREIT performance more accurately. Another limitation of the research is that it can only be said to hold for the US, meaning that it is low in external validity, to improve on this the research could be expanded to a more global level by including data from European, Asian or South American financial markets. Also, further research could include the ROE of MREITs to see how MREITs differ from EREITs in relation to market capitalization. A final limitation could be the lack of control variables, including more relevant control variables would improve the accuracy of the results, allowing for a better understanding of the relation between market capitalization and EREIT performance.
7. References
Bhuyan, R., Kuhle, J. L., Al-Deehani, T. M., & Mahmood, M. (2015). Portfolio diversification benefits using Real Estate Investment Trusts-An experiment with US common stocks, Equity Real Estate Investment Trusts, and Mortgage Real Estate Investment Trusts. International Journal of Economics and Financial Issues, 5(4).
Chen, K., & Tzang, D. (1988). Interest-rate sensitivity of real estate investment trusts.
Chiang, K. C., Lee, M. L., & Wisen, C. H. (2005). On the time-series properties of real estate investment trust betas. Real Estate Economics, 33(2), 381-396.
Delcoure, N., & Dickens, R. (2004). REIT and REOC systematic risk sensitivity. Journal of
Real Estate Research, 26(3), 237-254.
Gaille, B. (2018). 19 REIT Industry Industry Statistics and Trends. Retrieved from https://brandongaille.com/19-reit-industry-statistics-and-trends/
Garcia, V. F., & Liu, L. (1999). Macroeconomic determinants of stock market development.
Journal of Applied Economics, 2(1), 29-59.
Giliberto, M., & Shulman, D. (2017). On the interest rate sensitivity of REITs: evidence from twenty years of daily data. Journal of Real Estate Portfolio Management, 23(1), 7-20. Glascock, J. L., Lu, C., & So, R. W. (2000). Further evidence on the integration of REIT,
bond, and stock returns. The Journal of Real Estate Finance and Economics, 20(2), 177-194.
Kola, K., & Kodongo, O. (2017). Macroeconomic risks and REITs returns: A comparative analysis. Research in International Business and Finance, 42, 1228-1243.
Mulder, C. H. (2006). Population and housing: a two-sided relationship. Demographic
Research, 15, 401-412.
Odularu, O. (2009). The impact of share market capitalization on a companys performance: A case study in the Nigerian confectionary industry. African Journal of Business
Management, 3(5), 220-226.
Patel, R. C., & Olsen, R. A. (1984). Financial determinants of systematic risk in real estate investment trusts. Journal of Business Research, 12(4), 481-491.
Semer, S. L. (2009). A Brief History of US REITs. Can. Tax J., 57, 960.
Subeniotis, D. N., Papadopoulos, D. L., Tampakoudis, I. A., & Tampakoudi, A. (2011). How inflation, market capitalization, industrial production and the economic sentiment indicator affect the EU-12 stock markets.
Swanson, Z., Theis, J., & Casey, K. M. (2002). REIT risk premium sensitivity and interest rates. The Journal of Real Estate Finance and Economics, 24(3), 319-330.
