Does Financial Crisis Matter? Systematic Risk in the Casino Industry

Transcription

1 Does Financial Crisis Matter? Systematic Risk in the Casino Industry Dr. Day-Yang Liu, Professor, Graduate Institute of Finance, National Taiwan University of Science and Technology, Taiwan Cheng-Hsien Lin, Graduate Institute of Finance, National Taiwan University of Science and Technology, Taiwan ABSTRACT In December 2007, after a decade of rising market valuations and revenues, the U.S. casino industry began to suffer the effects of systematic risk caused by the 2008 financial crisis. Drawing data from 21 firms spanning , this study explores the relationship between systematic risk (beta) and six key financial variables for the U.S. casino industry: firm size, liabilities as a percentage of assets, asset turnover, return on assets, EBIT growth rate, and current ratio. We analyze the variables that most determine the industry s exposure to systematic risk and the variables that generate the most reliable predictions of beta as a measure of systematic risk. We find that only firm size significantly and affirmatively affects beta before, during, and after the 2008 financial crisis. Asset turnover and liabilities as a percentage of assets are significant predictors before and after the crisis. INTRODUCTION Many economists consider the financial crisis of , also known as the global financial crisis and 2008 financial crisis, to have been the worst financial crisis since the Great Depression of the 1930s. It threatened to collapse large financial institutions and was prevented by the bailout of banks by national governments, but stock markets still dropped worldwide. In many areas, the housing market also suffered, resulting in evictions, foreclosures, and prolonged unemployment. During the decade before 2007, the U.S. casino industry enjoyed significant growth in revenue and stock prices. However, performance began to decline as a crisis affecting the entire U.S. economy set in. The National Bureau of Economic Research (2008) dates the start of the crisis as December 2007 based on payroll employment, GDP, and gross domestic income, but the casino industry saw declines in business volumes starting in early According to the American Gaming Association (2013), casino market experienced fluctuations particularly from 2007 to 2013 (Fig. 1), the earliest year for which its website offers data. Although the industry s decline paralleled the nationwide economic downturn, no research has specified the factors that heightened its financial risk and eroded its stock prices. In 2009, the industry recorded a 5.36% fall in commercial casino revenue; however, in 2011, because of the pace of the U.S. economic recovery, U.S. commercial casinos rebounded. The Journal of Global Business Management Volume 11* Number 1 * April 2015 Issue 147

2 Figure 1: Gaming Revenue: 10-Year Trend. Sources: American Gaming Association Note: All amounts in billions USD Although systematic and unsystematic risks affect stock prices, diversified investors are concerned primarily with systematic risk, and they require higher rates of return for stocks of firms with heightened exposure to it. By understanding the influence of their decisions on financial ratios that affect exposure to systematic risk, casino executives can better manage the consequences of systematic risk on their firms stock prices. This study examines the determinants of casino companies systematic risk or beta. Drawing on earlier studies of other industries, the present study first identifies six financial variables that influence exposure to systematic risk. Then, it analyzes whether these financial variables predict a firm s systematic risk differently before and during a systematic crisis. LITERATURE REVIEW Unsystematic and systematic risks comprise the total risk to which prices of publicly traded stocks are exposed. The former is firm specific, but systematic risk beta includes broader market risk. Portfolio diversification diminishes investors vulnerability to unsystematic risk, leaving them concerned primarily with systematic risk, which is often determined by the Capital Asset Pricing Model (CAPM) theory (Sharpe, 1963); the theory states that: E(R i )=R f +β i (R m -R f ) (1) Where E(R i )= is the expected return on the capital asset E(R m )= is the expected return of the market R f =is the risk-free rate of interest such as interest arising from government bonds β i =systematic risk(the beta) Where R is the return on the i th security, Rm is the return on the market portfolio, e is the error regarding the regression line that represents the relationship between the two, β is the estimated beta of the i th security, and a is the estimated vertical intercept. The CAPM suggests that return on an asset is determined by adding the risk-free rate to a risk premium that increases as the company s exposure to 148 The Journal of Global Business Management Volume 11 * Number 1 * April 2015 Issue

3 systematic risk (beta) increases. This exposure can be estimated by the above-given equation and presented graphically(gu & Kim, 1998).To detect the influence of financial policies on systematic risk, different types of variables have been used in prior studies (Eldomiaty, 2009; Gu & Kim, 1998; Iqbal & Shah, 2012; Kim, Gu, & Mattila, 2002; Lee & Jang, 2007; Logue & Merville, 1972; Rowe & Kim, 2010). Large firms should have lower systematic risk due to economics of scale (Olibe, Michello, & Thorne, 2008). According to former researches the negative relationship has been found contended that in large companies systematic risk is low because the large firms have the ability to lesser the effect of economic changes(gu & Kim, 1998; Kim et al., 2002; Logue & Merville, 1972).Firm size is measured by taking the natural logarithm of total assets. Logarithm conversion condenses the effect of skewness. When managers increase the proportion of debt in their firms capital structures, the result is a greater risk exposure, expressed as a positive and nonlinear relation between leverage and systematic risk (beta). Found the beta as increasing function of leverage. Found the beta as increasing function of leverage. Frequent studies (Kim et al., 2002) hypothesized the positive relationship among leverage and beta. Debt ratio is used to calculate the leverage. Measured short term liabilities and long term liabilities separately because few firms use short-term liabilities indefinitely in their capital structures (Logue & Merville, 1972). Used leverage in their study as control variable and found positive relationship between leverage and systematic risk (Olibe et al., 2008). More operating efficiency means generating more profit and due to more profit the systematic risk is reduced (Kim et al., 2002). Generally researchers show the negative impact of operating efficiency on beta. Concluded the relationship of high efficiency and low systematic risk (Gu & Kim, 1998; Kim et al., 2002). They also found negative relationship in nonfinancial sectors between systematic risk and operating efficiency. Operating efficiency can be measured by asset turnover ratio (Eldomiaty, 2009). Success of any firm depends upon profitability and in profitable firms the chances of systematic risk reduce (Logue & Merville, 1972). Previous findings indicated a negative relationship between profitability and systematic risk (Iqbal & Shah, 2012; Kim et al., 2002; Lee & Jang, 2007; Rowe & Kim, 2010). However, in some particular industries this relation goes inversed. Beta is a diminishing function of growth (Hong & Sarkar, 2007). Rapid growth in companies increases systematic risk (Kim et al., 2002). Negative and positive relationship has been found among growth and systematic risk. Growth is positively related with systematic risk. Companies with high growth want more possessions or resources and for getting these resources firm need extra financing. Annual percentage change in earnings before interest and taxes is used to compute the growth of any firm. According to prior studies, liquidity has both positive and negative impact on systematic risk. Most studies concluded a negative relationship between systematic risk and liquidity (Eldomiaty, 2009; Gu & Kim, 1998; Iqbal & Shah, 2012; Kim et al., 2002; Lee & Jang, 2007; Logue & Merville, 1972). Found negative relationship among systematic risk (Beta) and liquidity. They argued that with increase in liquidity of the firm, the systematic risk decreases. However, Jensen (1984) disclosed a positive relationship among systematic risk and liquidity. He contended that with increase in liquidity agency cost of free cash flows of the firms also increase and this also increases systematic risk. Most investors use liquidity ratios at the time of investment to forecast the current position of any firm. Liquidity of the firm is calculated by current ratio. Argue that financial ratios that predict exposure to systematic risk vary by industry (Logue & Merville, 1972). Even so, the few extant studies of financial ratios and exposure to systematic risk are cursory and generally centered upon the hospitality industry. To determine which financial variables are The Journal of Global Business Management Volume 11* Number 1 * April 2015 Issue 149

4 potential determinants of beta for this study, previous studies have been evaluated in this section. Suggest that which financial variables predict systematic risk varies by industry (Logue & Merville, 1972). While minimal studies have been done, there are a few within multiple facets of the hospitality industry. Authors of the previous study concerning casinos, examined the period from 1992 to 1994 one of the industry s fast growth periods (Gu & Kim, 1998). Evaluating current ratios, leverage ratios, asset turnover, and profit margins of 35 firms, they found that only asset turnover significantly and negatively correlated with beta at the 10% level. All other variables were statistically insignificant. Furthermore, they found that stock prices for casino firms are more volatile than for equity market averages and suggested exploring additional liquidity, leverage, and profitability ratios for relationships between other ratios and beta. For example, they advised researchers to use quick ratios instead of current ratios because the former better represent a casino firm s liquidity. METHODOLOGY To simultaneously test hypotheses, we obtained data from 21 casino companies spanning We performed multiple linear regressions using beta as the dependent variable and six financial variables as the independent variables. These 21 firms were the only casino companies that were publicly traded, published public financial information, and owned and operated at least one physical casino during all ten years. Private equity firms, on-line casinos, casino management companies, or firms that had not disclosed financial information for any year were excluded. On the basis of literature review, six hypotheses are developed as follows: HI: Large casinos (Firm Size, US$ in billions) have high systematic risk. H2: Casinos with high leverage (Liabilities as % of Assets) have high systematic risk H3: Casinos with high efficiency (Assets Turnover Rate) will be subject to low systematic risk. H4: Profitability (Return on Assets) is negatively related to systematic risk H5: Casinos subject to fast growth (EBIT Growth Rate) have high systematic risk. H6: Liquidity (Current Ratio) have high systematic risk. In order to test these hypotheses simultaneously, a multiple linear regression analysis was performed with Beta as the dependent variable and the six financial variables as the independent variables in this study. Using the financial information of 21 public traded casino companies(see Appendix A for the list of companies), from 2004 to 2013, six financial variables were analyzed (See Appendix B for the list of variables). The 21 companies selected were the only casino companies that were publicly traded, had public financial information, and owned and operated at least one physical casino for all ten years. Any firm that has gone private equity, is an on-line casino, only manages casinos, or has yet to report any year's financial information was excluded. These variables were used as independent variables in multiple regression analysis. The regression equation developed in this study takes the following form: Beta=β 0 +β 1 (FS)+β 2 (LEV)+β 3 (OE)+β 4 (ROA)+β 5 (Growth)+β 6 (CR)+ε (2) Where; FS=Firm Size LEV=Leverage OE=Operating Efficiency ROA=Return on Asset Growth=EBIT growth rate CR= Current Ratio 150 The Journal of Global Business Management Volume 11 * Number 1 * April 2015 Issue

5 We took financial information from COMPUSTAT or from annual 10-K filings when COMPUSTAT data were unavailable. All ratios selected as variables were calculated using statistical software and were not extracted from the database. We used monthly stock returns, measured as a percentage of changes in prices, from the Wharton Research Data Services database spanning , which witnessed both faster growth and crisis of the casino industry. RESULTS Table 1 demonstrate the descriptive statistics of systematic risk (beta) and six independent variables for 21 listed firms for ten year period of Mean value of beta before, during and after the financial crisis are , and These mean value of beta are large than market beta that are always consider greater than 1 and also indicates that sample of listed firms are more riskier than market. The descriptive statistics including the mean, standard deviation, min and max for the 21 observations. Table 1: Descriptive Statistics Period/Variables Mean SD Min Max Before financial crisis Beta Firm Size(Natural Logarithm of assets) Leverage(Liabilities as % of assets) Operating Efficiency (Asset Turnover) Profitability (Return on Assets) Growth rate(ebit Growth Rate) Current ratio(cr) During crisis Beta Firm Size(Natural Logarithm of assets) Leverage(Liabilities as % of assets) Operating Efficiency (Asset Turnover) Profitability (Return on Assets) Growth rate(ebit Growth Rate) Current ratio(cr) After financial crisis Beta Firm Size(Natural Logarithm of assets) Leverage(Liabilities as % of assets) Operating Efficiency (Asset Turnover) Profitability (Return on Assets) Growth rate(ebit Growth Rate) Current ratio(cr) Table 2 show we checked for multicollinearity among variance inflation factors (VIF). No formal criterion sets a definitive standard for either. Some argue that a tolerance value below 0.1 or VIF exceeding 10 roughly reflects significant multicollinearity. Others cite a conditioning index exceeding 30 for a given dimension coupled with variance proportions exceeding 0.5 for at least two different variables. Table 2 reveals no multicollinearity. In three models (before, during, and after), all VIF are below 10. The magnitude of the partial regression coefficient depends, among others, on the units in which the variable is measured (e.g., Size= US$ in billions; Leverage= %). To make the partial regression coefficients The Journal of Global Business Management Volume 11* Number 1 * April 2015 Issue 151

6 more comparable, this study used standardized coefficients (Z score). The positive standardized coefficient (beta of.339 indicates that there was a statistically significant (p <.05) linear relationship between efficiency (measured by Firm size) and a casino company's systematic risk (measured by Beta) before financial crisis. A casino company's leverage also showed a significant linear relationship with the company's systematic risk (Leverage =0.254, t =1.804, p <.1). The negative standardized coefficient (beta of indicates that there was a statistically significant (p <.05) linear relationship between efficiency (measured by asset turnover ratio). During crisis indicates the magnitude of each financial variable related to beta separately. The positive standardized coefficient (β) of.52 confirms statistical significance (p < 5%). Size again significantly affects its exposure to systematic risk (t = 2.803, p <.01). For every positive degree of increase in firm size, predicted beta increases.992. The other five financial variables are insignificant. After crisis indicates separate magnitudes of each financial variable related to beta. Casino company's size also showed a significant linear relationship with the company's systematic risk. The positive standardized coefficient beta of.188 indicates both that there was a statistically significant (p <.05) linear relationship between efficiency measured by Leverage) and there exists among casino companies systematic risk (measured by Beta) after financial crisis. The negative standardized coefficient beta of indicates that there was a statistically significant (p <.05) linear relationship between efficiency (measured by asset turnover ratio). On the other hand, there were insignificant associations between the other financial variables and Beta. Table 2: Coefficients of Financial Variables Unstandardized Standardized Coefficients Coefficients Correlations Collinearity Statistics B Std. Error Beta t Sig Zero-order VIF I.Before crisis (Constant) Firm Size(FS) Leverage(LEV) Operating Efficiency (OE) Profitability (ROE) Growth Current ratio(cr) II.During crisis (Constant) Firm Size(FS) Leverage(LEV) Operating Efficiency (OE) Profitability (PROF) Growth Current ratio(cr) III.After crisis (Constant) Firm Size(FS) Leverage(LEV) Operating Efficiency (OE) Profitability (PROF) Growth Current ratio(cr) The Journal of Global Business Management Volume 11 * Number 1 * April 2015 Issue

7 As Table 3 indicates, the pre-crisis ( ) absolute value of the correlation coefficient between all six factors and beta is From the regression model, 44.2% of the variation in beta (systematic risk) is explained by variation in the six financial variables. This result confirms the model s significance: F = 4.205, p < 0.1. The relationship between the six financial variables and beta is linear. Table 3 during crisis provides descriptive statistics, including means and standard deviations, for the 21 observations. The absolute value of the era correlation coefficient between all six factors and beta is That is, 44.8% of the variation in beta (systematic risk) is explained by variation in the six financial variables. Results confirm the model s significance: F = 4.658, p < The linear relationship between the six variables and beta during the period is strong. Table 3 indicates descriptive statistics, including means and standard deviations, for the 21 observations. The absolute value of the era correlation coefficient between all six factors and the beta is That is, variations in the six financial variables explain 58.3% of the variation in beta (systematic risk). Results again confirm the model s significance: F = , p < The linear relationship between the six variables and beta during is strong. Table 3: Model Summary Model R R Square Std. Error of the Estimate F Sig Before During After Table 4 compares this study s results with its hypotheses. Firm size is the only variable that significantly and positively correlates to beta before, during, and after the financial crisis. This finding defies the predicted correlation in this study and that in previous studies. Perhaps, inconsistencies arise because casino companies added properties very quickly during , exacerbating competition and market saturation and elevating chances of bankruptcy. Consequently, firms may have been highly vulnerable to default risk. Confirmatory investigation is needed. Liabilities (as a percentage of assets) significantly correlate with beta before and after the crisis. This finding supports previous findings that higher leverage engenders higher risk because shareholders realize that their claims on earnings have been subordinated further. This finding also supports Gu (1993) and Rowe (2010) suggestion that hospitality firms and the gaming industry are sensitive to economic downturns. Our findings emphasize that casino companies must manage debt and reduce associated financial risk, especially during non-financial crises. Asset turnover significantly determined beta before and after the financial crisis of , a finding consistent with previous studies. However, we found that asset turnover correlates negatively but without statistical significance during the crisis. This result indicates that a higher asset turnover is associated with higher risk outside times of crisis. Gu (1993) and Rowe (2010) postulate that a negative correlation between asset turnover and beta indicates that efficient asset management can engender lower systematic risk for casino companies. Our contradictory finding implies that casino companies had achieved high asset efficiency before the crisis, but doing so did not reduce their vulnerability to systematic risk, perhaps because the industry had enlarged capacity rapidly into a saturated market. The implication is that financial managers should control both operating ratios (asset turnover) and ratios related to financing and investing. The Journal of Global Business Management Volume 11* Number 1 * April 2015 Issue 153

8 Table 4: Hypotheses Results Financial Variables Before crisis During crisis After crisis H1 Firm Size(FS) Positive* Positive** Positive* H2 Leverage(LEV) Positive* Positive Positive* H3 Operating Efficiency (OE) Negative * Negative Negative * H4 Profitability (PROF) Positive Negative Negative H5 Growth rate (EBIT Growth Rate) Positive Positive Negative H6 Current ratio(cr) Negative Positive Positive *p<.05; **p<.01 CONCLUSION Drawing data from 21 firms spanning , this study explores the relationship between systematic risk (beta) and six key financial variables for the U.S. casino industry. Scholars also could use data from a complete sample set and not from subsamples. In all previous studies, relatively small sample size is a concern because few casino firms trade publicly. Future research could include more companies by increasing the span of years studied and by not averaging data, although doing so might generate results influenced by firms entering, exiting, or merging rather than by the financial factors being evaluated. We find that only firm size significantly and affirmatively affects beta before, during, and after the 2008 financial crisis. Asset turnover and liabilities as a percentage of assets are significant predictors before and after the crisis. The study s limitations include its small time span and sample size, both of which are beyond our control. Future research could include more years and dummy variables for periods before, during, and after the crisis. REFERENCES Association, A. G. (2013). Gaming revenue: 10-year trend. Online: org/industry-resources/research/fact-sheets/gaming-revenue-10-year-trends. Eldomiaty, T. I. (2009). The fundamental determinants of systematic risk and financial transparency in the DFM General Index. Middle Eastern Finance and Economics(5). Gu, Z. (1993). Debt use and profitability: a reality check for the restaurant industry. Foodservice Research International, 7(3), Gu, Z., & Kim, H. (1998). Casino firms' risk features and their beta determinants. Progress in Tourism and Hospitality research, 4(4), Hong, G., & Sarkar, S. (2007). Equity Systematic Risk (Beta) and Its Determinants. Contemporary Accounting Research, 24(2), Iqbal, M. J., & Shah, S. Z. A. (2012). Determinants of systematic risk. The Journal of Commerce, 4(1), 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), Lee, J.-S., & Jang, S. S. (2007). The systematic-risk determinants of the US airline industry. Tourism Management, 28(2), Logue, D. E., & Merville, L. J. (1972). Financial policy and market expectations. Financial Management, National Bureau of Economic Research. Business Cycle Dating Committee. (2008). Determination of the December 2007 Peak in Economic Activity. Retrieved August 19, 2009 from Olibe, K. O., Michello, F. A., & Thorne, J. (2008). Systematic risk and international diversification: An empirical perspective. International Review of Financial Analysis, 17(4), Rowe, T., & Kim, J. S. (2010). Analyzing the relationship between systematic risk and financial variables in the casino industry. UNLV Gaming Research & Review Journal, 14(2), Sharpe, W. F. (1963). A simplified model for portfolio analysis. Management science, 9(2), The Journal of Global Business Management Volume 11 * Number 1 * April 2015 Issue

9 Appendix A: List of Firms 1. Affinity Gaming 2. Boyd Gaming 3. Caesars Entertainment 4. Canterbury Park Holding 5. Century Casinos 6. Dover Downs Gaming & Entertainment 7. Empire Resorts 8. Full House Resorts 9. Isle of Capri Casinos 10. Lakes Entertainment 11. Las Vegas Sands 12. Melco Crown Entertainment 13. MGM Resorts International 14. Mohegan Tribal Gaming Authority 15. Monarch Casino & Resort 16. MTR Gaming Group 17. Nevada Gold & Casinos 18. Penn National Gaming 19. Pinnacle Entertainment 20. Trans World 21. Wynn Resorts Appendix B: List of Variable Name of Variable Firm Size(FS) Leverage(LEV) Operating Efficiency (OE) Profitability (PROF) Growth Current ratio(cr) Independent Variables Measurement LN(Total Asset) Debt ratio = Total Debt / Total Assets Asset Turnover = Total revenue / Total Asset Return on Assets = Net income / Total Assets Percentage change in earnings before interest and taxes Current asset /Current liabilities The Journal of Global Business Management Volume 11* Number 1 * April 2015 Issue 155