Understanding corporate philanthropy in the hospitality industry

International Journal of Hospitality Management 48 (2015) 150–160

Contents lists available at ScienceDirect

International Journal of Hospitality Management journal homepage: www.elsevier.com/locate/ijhosman

Understanding corporate philanthropy in the hospitality industry Ming-Hsiang Chen a,b,1 , Chien-Pang Lin c,∗ a

Department of Finance, National Chung Cheng University, Chia-Yi, Taiwan, ROC School of Hospitality Business Management, Washington State University, Pullman, WA 99164-4742, USA c Department of Finance, Chang Jung Christian University, Tainan, Taiwan, ROC b

a r t i c l e

i n f o

Keywords: Corporate philanthropy Value enhancement theory Agency cost theory Hospitality industry

a b s t r a c t This study makes a unique contribution to the corporate social responsibility literature, both from a theoretical perspective and corporate philanthropy (CP) perspective that has been under-researched so far. Specifically, it extends the investigation of the motives of CP to the hospitality industry in the emerging country of Taiwan and is the first examination of the determinants of CP in the hospitality industry based on the value enhancement (VE) and agency cost (AC) theories. Previous studies found that either AC or VE determinants, not both, play a major role in making philanthropy decisions. This study reveal new and interesting findings using the data from 13 publicly traded hospitality companies in Taiwan collected between 1996 and 2011. Panel regression test results show that both VE determinants (labor intensity and profitability) and AC determinants (board size and the ratio of independent directors to total directors) were found to be significant factors in explaining the discretionary behavior of CP in the hospitality industry. Furthermore, the lagged philanthropy-to-sale ratio was also significantly related to CP decisions in the hospitality industry in Taiwan, suggesting that the behavior of CP in the Taiwanese hospitality industry is persistent. The possible explanation is that changing philanthropy pattern may lead to market-adverse expectations and reactions and damage the reputation of companies. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Corporate philanthropy (CP) is the discretionary act of corporations donating a portion of their profits or resources to nonprofit organizations. An interesting question is why corporations give to charity. Fry et al. (1982), Amato and Amato (2012), and Du et al. (2013) stated that empirical questions about CP were limited to surveys of corporate executives, making it difficult to distinguish what corporations actually do from what their executives claim that these corporations do. As Fry et al. (1982) noted, although the rationale for CP could be altruism, CP data offer little insight into the altruistic motive of CP if CP is motivated primarily by profit-related considerations. Academic studies have examined the reasons for corporations’ philanthropic contributions. Several studies (Brown et al., 2006; Fry et al., 1982; Navarro, 1988) have focused on the profit-maximizing or self-motivated managerial perquisite motives instead of on altruism. The profit motivation argument is examined here based

∗ Corresponding author. Tel.: +886 62785123. E-mail addresses: [email protected] (M.-H. Chen), [email protected] (C.-P. Lin). 1 Tel.: +886 52720411; fax: +886 52720818. http://dx.doi.org/10.1016/j.ijhm.2015.04.001 0278-4319/© 2015 Elsevier Ltd. All rights reserved.

on value enhancement theory and the argument of self-motivated managerial perquisite is explained by the agency cost theory. Value enhancement theory postulates that CP creates value for shareholders. Brown et al. (2006) argued that CP may be regarded as a form of moral capital investment, which could enhance a firm’s image and reputation among its stakeholders. This positive image increases firm value by promoting customer loyalty (Lev et al., 2010), motivating employees (Greening and Turban, 2000; Navarro, 1988), inducing preferential treatment from regulators or policymakers (Brown et al., 2006), or providing insurance-like protection against risk (Godfrey, 2005). In this regard, value enhancement theory of CP could be considered a specific form of stakeholder theory, that is, a model of analysis, which addresses the relationship between both moral and value objectives of a firm and its stakeholders (Roberts, 1992). Agency cost theory proposes that managerial insiders who participate in CP may increase their own utility through CP while shareholders incur expenses. CP may provide managerial insiders with philanthropy-related benefits and offer an opportunity to support their own pet charities to enhance their personal reputation (Brown et al., 2006). Managerial insiders thus may have the incentive to pursue their own interests through CP. As the cost of philanthropy has exceeded its financial contributions, CP represents an agency cost because it satisfies the managerial insiders’

M.-H. Chen, C.-P. Lin / International Journal of Hospitality Management 48 (2015) 150–160

151

Fig. 1. The average corporate philanthropy rate.

taste for “doing good” while shareholders suffer an opportunity loss (Brown et al., 2006). However, empirical research on whether value enhancement theory or agency cost theory better explains CP has been mixed and inconclusive (Brown et al., 2006; Fry et al., 1982; Navarro, 1988). As Navarro (1988) and Brammer and Millington (2004) noted, empirical studies on the motives driving CP have been limited by the unavailability of published information or data on CP. Because of limited data, most previous studies on driving motives of CP have been conducted in developed countries, such as the UK and USA (Zhang et al., 2010). So far, no research has examined whether agency cost, value enhancement, or other determinants can more accurately explain the CP decision in an emerging country. Since differences between developed and emerging countries in firm characteristics, agency conflicts, and industry environment that are due to cultural, social, economic and political factors are expected (Wright et al., 2005; Young et al., 2008), it is plausible that the determinants of CP in an emerging country would be different from those in developed countries. At the same time, the determinants used to explain CP vary across industries (Amato and Amato, 2012; Brammer and Millington, 2008; Brown et al., 2006). Further, no hospitality research has analyzed the CP issue. Note that while Turner et al. (2001) presented three ways in which charities in tourism in the UK are involved and the implications this may have for the industry, they did not examine the motives of charities in tourism. This paper is the first empirical research study to test and explain the rationales for CP in the hospitality industry, and the availability of CP data in Taiwan enables us to examine the issue in the context of an emerging country. According to the data taken from Market Observation Post System (http://newmopsov.twse.com.tw/), the value of CP made by Taiwanese publicly traded companies amounted to approximately 2.797 billion NT$ (New Taiwan dollars) in 2011. This figure represents about .282% of total annual pre-taxable profits of all publicly

traded companies in Taiwan. The total value of CP from hospitality companies amounted to about NT$ 11.619 million in 2011. Among 27 industry sectors listed on the Taiwan Stock Exchange from 1996 to 2011, the hospitality industry was ranked second in terms of the ratio of the total value of CP to sales revenue, following the biotechnology and medical care industry (see Fig. 1). This suggests that hospitality companies in Taiwan were eager to donate a higher percentage of their sales revenue compared to other industry sectors. Moreover, Liu et al. (2006) stated that the power of managers in Taiwanese firms is quite high. However, with a high power in the companies, managers tend to exercise significant control over the firm, and the agency problems can become problematic for shareholders. This principal–agent relationship resulting from ownership structures may have an influence on CP decision. Given the illustration above, it would be interesting to examine the determinants of the enthusiastic CP behavior of the Taiwanese hospitality publicly traded companies. Thus, this study makes a unique contribution to the corporate social responsibility (CSR) literature from both theoretical and corporate philanthropy perspectives. The remainder of this study is organized as follows. Section 2 reviews the literature. Data and the variables are presented in Section 3. Section 4 reports methodology and empirical results. Discussion of the findings and conclusion are presented in Section 5. 2. Literature review While researchers in the management, finance, and other fields are interested in the issue of CP, only a few empirical studies have examined the motives of CP given the limited data available. Brown et al. (2006) stated that CP motivated by value enhancement might boost corporate future revenue growth while philanthropic donation driven by agency cost may cause the managerial insiders to pursue their own interests. As a result, discretional philanthropy

152

M.-H. Chen, C.-P. Lin / International Journal of Hospitality Management 48 (2015) 150–160

may be donated to maximize either firm value or the utility of the managerial insiders. The studies that have utilized agency cost and value enhancement theories are reviewed below. Value enhancement theory argues that CP enhances firm value. Fry et al. (1982) tested the profit motivation argument of CP using the data from several US industry sectors. They suggested that CP complements advertising and is therefore a profit-maximizing expense. Similarly, Brammer and Millington (2008), Navarro (1988), Wang et al. (2008), and others also confirmed that firms with higher advertising expenses were more likely to make larger philanthropic donations. On the contrary, Amato and Amato (2012) found no significant effect of advertising expenses on retail CP in the US, but they did find that firm size and industry effect significantly explained the retail CP behavior. Navarro (1988) examined the profit maximization and managerial utility maximization motives that drive CP using a pooled cross-section of 249 large US firms. He found that CP was driven primarily by profit-maximizing rather than managerial utilitymaximizing and factors such as advertising expenses and labor intensity. Thus, profit maximization appears to be an important motive of CP. Zhang et al. (2010) found that CP is significantly associated with advertising intensity and profitability. They also found that firms in competitive industries contributed more cash to philanthropy. The findings thus provided evidence for value enhancement motive, showing that CP is used as a business strategy to improve firm performance. According to the explanation of agency cost, managerial insiders increase their own utility through philanthropy contributions. Agency cost is significantly related to the ownership structure (Jensen and Meckling, 1976). Separation of ownership and control yields significant power to managerial insiders. Managerial insiders, including managers and directors, have different roles and responsibilities. While the ultimate responsibility of managers is to maximize shareholders’ value, directors’ role is to oversee and evaluate the management’s work and to assure that the management’s interests are aligned with those of the shareholders (Ross et al., 2007). The lack of oversight by a board of directors could result in serious agency problems (Fama and Jensen, 1983). Furthermore, larger boards tend to become less effective in monitoring because larger boards may increase free rider and agency problems (Jensen, 1993). Thus, as the board size increases, the board may have an incentive to maximize its own utility. Therefore, a positive link between board size and agency conflicts may arise. Overall, if CP is an agency cost, not only managers, but also all board members could benefit from their firm’s CP. Agency cost may also be related to concentrated ownership, such as institutional owners. Powerful institutional owners are more likely to curtail managerial excesses to protect owners’ interests (Agrawal and Knoeber, 1996). Thus, stronger institutional ownership is more likely to exercise effective monitoring, resulting in lower agency cost. If a firm’s institutional owners perceive CP as excessive, more institutional ownership would strengthen the monitoring of managerial discretion and hence the firm would be less likely to contribute to philanthropy. In addition, debt may represent an agency cost (Jensen, 1986). Firms with higher debt leverage are more likely to face the risk of bankruptcy or liquidation and devote their resources to minimizing the risk of the negative financial consequences of debt leverage (Adams and Hardwick, 1988). Brown et al. (2006) stated that high debt is likely to achieve higher monitoring efficiency and thus decrease monitoring cost of a company. Accordingly, firms with higher debt leverage are less likely to participate in CP, and more likely to fall short of discretionary profits and achieve higher monitoring efficiency and. Adams and Hardwick (1988) examined the determinants of the level of CP using a sample of 100 publicly traded UK companies.

They found no association between CP and ownership concentrations but found that CP was positively related to firm size and profitability and negatively related to debt leverage. Brammer and Millington (2004) studied the development of the UK CP based on 416 publicly traded UK manufacturing and service companies. They found that CP was related to profits, debt leverage, and firm size, but they found no connection between CP and shareholder dispersion. Adams and Hardwick (1988) and Brammer and Millington (2004) concluded that CP was not significantly associated with the separation of ownership. In contrast, Bartkus et al. (2002) compared big and small givers using a matched-paired US sample and found that influential institutional owners perceived CP as excessive and tried to limit it. Brown et al. (2006) scrutinized the CP practices using a sample of 241 Fortune 500 firms. They suggested that agency cost motives rather than value enhancement motives drive CP. Specifically, institutional ownership and the ratio of inside directors to total directors had no significant effect on philanthropic contributions. However, firms with a larger board and lower debt leverage were more likely to provide philanthropy contributions. On the other hand, Wang and Coffey (1992) found a positive and significant link between CP and the ratio of insiders to outsiders based on a sample of 78 Fortune 500 firms. Although previous studies have generated some valuable and insightful findings, they have some limitations. First, it is not clear whether CP donations are made with the consent of the firm’s owners or whether they reflect self-aggrandizing behavior of the management (Hart, 1993). Empirical research on the role of value enhancement or agency cost theory in explaining CP remains ambiguous. Past studies, as reviewed in this section, tended to examine the influences of some individual value enhancement or agency cost determinants on CP, and the explanatory power of the determinants on CP found in these studies seems to be small. Instead of using some selected determinants of CP, this study provides a more comprehensive investigation by using a broader set of both value enhancement and agency cost determinants to test the hospitality philanthropy behavior. Second, due to data limitation, the use of cross-sectional data (for example, Bartkus et al., 2002; Brammer and Millington, 2004; Wang and Coffey, 1992; Zhang et al., 2010) failed to capture the possible fluctuations in the annual data of philanthropic contributions. This study used a longitudinal data set of the determinants of CP within an emerging country, which can offer a comparative analysis. Moreover, given the differences in public contact as well as in public relations exposure, firms from different sectors are expected to adopt different approaches to philanthropic donation (Amato and Amato, 2012). As a result, determinants used to explain the rationale for CP vary across industries and countries. Further, high level of data aggregation across broad industries would produce estimation bias. This study used a sample of hospitality firms with similar public contact to examine the motives that drive CP. This may provide unbiased estimation and help understand the behavior of CP in the hospitality industry. Third, the traditional ordinary least square (OLS) method was used to test the determinants driving CP (Adams and Hardwick, 1988; Amato and Amato, 2012; Brammer and Millington, 2004; Brown et al., 2006; Wang and Coffey, 1992; Zhang et al., 2010). However, when heteroskedasticity and/or autocorrelation exist in the error terms, the residuals from OLS regression usually fail to satisfy the basic assumptions of the linear regression. Consequently, the regression test results could be biased. This study used panel regression tests to investigate the determinants of CP in the hospitality industry. Hsiao (1986) and Baltagi (2005) argued that panel data methodology can control for individual firm’s heterogeneity, reduce problems associated with multicollinearity and estimation

M.-H. Chen, C.-P. Lin / International Journal of Hospitality Management 48 (2015) 150–160

153

bias, and specify the time-varying relation between dependent and independent variables.

determinants in this study and calculated as employee expenses (EMPE) divided by sales revenue:

3. Data, variables and hypotheses

LINT =

Based on the classification of the Taiwan Stock Exchange, this paper uses a study sample that covers 13 publicly traded hospitality companies: Ambassador Hotel, China Airlines, EVA Airways Corporation, First Hotel, Grant Formosa Regent Taipei, Hotel Holiday Garden, Hotel Royal Chihpen, Landis Taipei Hotel, Leofoo Hotel, Janfusun Fancy World, Wanhwa Enterprise Company, Phoenix Tours International, Inc. and New Palace International Company Ltd. Obtained from the Market Observation Post System (http:// newmopsov.twse.com.tw/), the annual CP data are available from 1996 to 2011 for each sample hospitality company. All annual accounting or financial data of hospitality companies over the entire sample period are retrieved from the financial database of the Taiwan Economic Journal. Accordingly, there are 208 annual observations available for the panel regression tests. The dependent variable is CS, computed as the ratio of the total value of CP (TVCP) to sales revenue: CS =

TVCP . Sales revenue

(1)

Independent or explanatory variables (i.e. value enhancement and agency cost determinants) and control variables are defined and explained as follows. 3.1. Value enhancement determinants The variable of advertising rate (ADR) is measured as advertising expenses (ADVE) divided by sales revenue: ADR =

ADVE . Sales revenue

(2)

Roberts (1992) stated that corporate philanthropic activities could be regarded as a form of corporate strategy to advertising public image of social responsibility. Fry et al. (1982) indicated that CP may be a complement to advertising and thus is a rational business expenses to maximize shareholders’ value. CP should rise (fall) with the increase (decrease) of advertising expense rate. Empirical studies found a positive link between advertising expenses and CG (Fry et al., 1982; Navarro, 1988; Zhang et al., 2010). Hypothesis I. There is a positive relationship between advertising expense rate and CP. The profitability variable is measured by return on assets (ROA), which is defined as net income divided by total assets: ROA =

Net income . Total assets

(3)

Return on assets measures the ability of a firm’s management to generate profits from the firm’s assets. A high return on assets is an indication of efficient management. Roberts (1992) and Ullmann (1985) illustrated a significantly positive link between profitability and corporate social behavior. The reason is that firms with higher profitability because of their excellent management may have more discretionary profits to contribute to socially responsible activities. Thus, more profitable firms are likely to have more discretionary funds to contribute to philanthropy. Hypothesis II. A positive link between corporate profitability and philanthropy is expected. Following Brammer and Millington (2008), the variable of labor intensity LINT is used as one of the value enhancement

EMPE . Sales revenue

(4)

Many of the benefits of CP are likely to accrue to labor rather than to capital (Navarro, 1988). To attract, retain and motivate employees, firms may use CP as a management strategy to increase their moral and production efficiency. Accordingly, firms with a higher level of labor intensity are expected to be more involved in CP. Thus, a positive relationship between CG and labor intensity is assumed. Hypothesis III. ciated with CP.

Labor intensity is expected to be positively asso-

The R&D expenses rate (RDR) is used to measure the firm’s intangible investment per dollar of sale and calculated as R&D expenses (RDE) divided by sales revenue: RDR =

RDE . Sales revenue

(5)

When a firm invests intensively in research and development, intangible assets will be larger (Brown et al., 2006; Cowan et al., 2013). Such intangible assets could make the firms vulnerable to lawsuits and governmental regulations. Since public displays of CSR may be used by management as a strategy designed to satisfy general public and government demands (Roberts, 1992), firms may have value-enhancing incentives to seek protection through CP by creating goodwill with potential jurors, judges and regulators. The literature supports that firms with higher labor intensity give significant amounts of cash to charity (Brammer and Millington, 2004; Brown et al., 2006; Cowan et al., 2013; Wang et al., 2008). Hypothesis IV.

There is a positive link between RDR and CP.

3.2. Agency cost determinants The agency cost determinants of CP used in this study are debt leverage, the board size, the ratio of independent directors to total directors, managerial insider shareholdings and institution shareholdings. The variable of debt leverage (DLEV) is measured as total debts divided by total assets: DLEV =

Total debt . Total assets

(6)

Debt leverage acts as an indicator of a firm’s financial ability to meet its debt obligations. It controls management excesses by constraining the manager’s ability to spend corporate funds, reducing free cash flow and thus decreasing opportunities to misdirect funds (Jensen, 1986). Accordingly, high debt leverage is likely to result in a company’s higher monitoring efficiency and lower monitoring costs (Brown et al., 2006; Navarro, 1988). If CP acts as an agency cost, then firms with rigorous outside monitoring should give less to charity. Thus, CP rate should rise as leverage falls, and fall as leverage rises (Adams and Hardwick, 1988; Brown et al., 2006; Cowan et al., 2013). Hypothesis V. to CP.

Debt leverage is expected to be negatively related

Board size (BS) is calculated as the number of the board of directors. Jensen (1993) argued that as board size increases, the managerial process involves more social interactions. The larger the board, the more often the free-rider and agency problem may arise, as the directors may pursue their own interests without aligning them with those of shareholders. Hence, as board size increases, the board discretionally tends to donate more to philanthropy (Brown et al., 2006).

154

M.-H. Chen, C.-P. Lin / International Journal of Hospitality Management 48 (2015) 150–160

Hypothesis VI. A positive relationship between board size and CP is anticipated. In addition to the board size, the study considers the impact of the ratio of independent directors to total directors. Independent (or outside) directors have no material or pecuniary relationship with company or related persons (except sitting fees) and do not own company stock. While the primary role of the board of directors is to monitor corporate management, Chen et al. (2013) and Chou et al. (2013) argued that in Taiwan, independent directors provide more rigorous oversight than do non-independent (or inside) directors. It is commonly assumed that companies with a larger number of independent directors are associated with better corporate governance and/or firm performance (Chen et al., 2013). Consequently, independent directors are more likely than dependent directors to mitigate agency problems through their monitoring role in Taiwan’s publicly traded firms (Chou et al., 2013). The ratio of independent directors to total directors (INDBS) is given as: INDBS =

INDIR . BS

(7)

where INDIR denotes the number of independent directors. Since independent directors tend to do a better job of ensuring that the managerial insiders’ interests are aligned with those of shareholders, it is more likely that independent directors would prevent managerial insiders from advancing their own interests by CP at the expense of the shareholders. Hypothesis VII. CP is expected to be negatively related to the ratio of independent directors to total directors. Further, managerial insider shareholdings are an effective incentive that drives managerial insiders to align managerial interests with shareholder value to enhance corporate performance. Managerial insider shareholdings (INSIDE) is measured as the percentage of total shares held by directors, managers and stockholders who own at least 10% of the firm’s shares. As managerial insiders own a higher fraction of a firm’s shares, the managerial insiders have the incentive to enhance the revenue growth and thus may concentrate on corporate performance by maximizing profit, instead of making CP. Hypothesis VIII. There is a negative association between managerial insider shareholdings and CP. The variable of institution shareholders also performs a function to monitor managers and protect shareholder’s value. The variable of institution shareholdings (INSTIT) is defined as the ratio of shares held by institutional investors (SHI) to the total shares outstanding (TSO). INSTIT =

SHI . TSO

(8)

The more concentrated the ownership, the higher the monitoring efficiency that is exercised on managerial discretion (Brown et al., 2006). Firm with higher institutional holdings are less likely to participate in CP. Hypothesis IX. There is a negative link between institutional shareholdings and CP. 3.3. Control variables To control for possible firm size effect on CP, firm size (SIZE) is measured as the logarithm of the total assets. Roberts (1992) argued that large firms are under greater public and political scrutiny than small firms and hence may promote a socially conscious image to

induce friendlier treatment by customers, regulators and policymakers. Adams and Hardwick (1988) and Brammer and Millington (2004) also suggested that larger firms are more likely to engage more in CP. Moreover, Chen (2010) pointed out that large Taiwanese hospitality companies are likely to have a better market power and a greater ability to adapt to economic and social changes in a competitive environment and thus have a better corporate performance. Therefore, large hospitality companies may have more discretionary profits for CP. Firm age (AGE), proxied for firm’s maturity, is measured as the number of years since the inception of the firm. As a corporation matures, its social status and history of social involvement in philanthropic supports can become strengthened (Roberts, 1992). Any significant reduction of charitable sponsorship could be a signal that the corporation is likely to confront financial or managerial disorder, which may cost the corporation considerably. Thus, the firm tends to continue its philanthropy to prevent the risk of negative social expectations. Therefore, firm age is expected to be positively related to CP (Brown et al., 2006). The one-year lagged CP-to-sales ratio (CS(−1)) is used as an explanatory and a controlling variable to capture the persistent corporate behavior of involvement in CP. According to Lev et al. (2010), the lagged philanthropy-to-sale ratio offers insights into the behavior of CP. Similar to firm age, the lagged philanthropy-to-sale ratio acts as a firm’s indicator of tendency to make CP. Hence, the firms that donated to philanthropy last year are likely to continue this year, i.e. lagged philanthropy-to-sale ratio is positively related to CP. In addition, Lee and Yeh (2004) and Liu et al. (2006) indicated that publicly traded companies in Taiwan are generally controlled by families. Yeh et al. (2001) reported that 51.44% of the publicly traded Taiwanese companies are family controlled. Lee and Yeh (2004) stated that the controlling family owned 27.72% of the voting rights and family members occupied 59.98% of the board seats. As a result, the controlling family in the publicly traded Taiwanese companies is considerably powerful. To control for its potential influence on CP decision, the family-controlled hospitality firm is also used as a control variable. According to the Taiwan Economic Journal database, a firm is classified as family controlled (FAMILY) if controlling family members occupy at least 33% of the board seats or the family has at least 20% of the voting rights. The dummy variable FAMILY takes on a value of 1 if the hospitality company is controlled by family and 0 otherwise. 3.4. Summary statistics Table 1 reports the summary statistics of CP rate and determinants of CP of publicly traded hospitality companies in Taiwan. CS ranges from .000 to .022 with a mean of 0.477 × 10−3 and a standard deviation of .002. Among the value enhancement determinants, the sample mean of ADVR advertising expense ratio is .013, ranging from .000 to .104, suggesting that the sample hospitality companies have different advertising strategies to promote sales. Ranging from −12.540 to 17.740, PROT has a highest sample mean of 3.805 and a highest standard deviation of 5.146. The results indicate the various values of profitability across the sample hospitality companies. The variable of LINT varies from .018 to .451 with a mean of .150 and a standard deviation of .131. In addition, ranging from .000 to 0.380 × 10−3 , RDR has a very low mean of 0.103 × 10−4 with a standard deviation of 0.529 × 10−4 . The results indeed confirm that the Taiwanese hospitality industry is not a R&D intensive industry. The agency cost determinant DLEV fluctuates between .076 and .859 with a sample mean of .356 and a standard deviation of .195, showing the various structures of debt management across the sample hospitality companies. The sample mean of BS is 6.817, ranging from 3 to 23. Ranging from .000 to .429, INDBS has a sample

M.-H. Chen, C.-P. Lin / International Journal of Hospitality Management 48 (2015) 150–160

155

Table 1 Summary statistics of corporate philanthropy rate and philanthropy determinants. Variable

Mean

Dependent variables CS

Median

Maximum

Minimum

Standard deviation

.477 × 10−3

.000

Value enhancement determinants .013 ADVR 3.805 PROT .150 LINT .103 × 10−4 RDR

.005 3.550 .113 .000

.104 17.740 .451 .380 × 10−3

Agency cost determinants DLEV BS INDBS INSIDE INSITIT

.295 5.000 .000 .564 .534

.859 23.000 .429 .869 .853

.076 3.000 .000 .157 .089

.195 3.491 .091 .154 .241

.000 15.464 36.000 1

.022 19.317 54.000 1

.000 12.671 9.000 0

.002 1.702 12.479 .463

Control variables CS(−1) SIZE AGE FAMILY

.356 6.817 .030 .541 .504 .483 × 10−3 15.516 33.831 .692

.022

mean of .030 and a standard deviation of .091, illustrating that the majority of board size of the Taiwanese hospitality companies are non-independent directors. INSIDE ranges from .157 to .869 with a mean of .541 and a standard deviation of .154. The results imply that the degree of unification between control and management and the voting power of the managerial insiders are quite high for the sample hospitality companies. In addition, ranging from .089 to .853, INSITIT has a mean of .504 with a standard deviation of .241, indicating the high voting power of the institution shareholders on sample hospitality companies. For the control variables, ranging from .000 to .022, the lagged philanthropy-to-sale ratio CS(−1) has a mean of 0.483 × 10−3 and a standard deviation of .002. Moreover, hospitality companies of different sizes are included in the sample. The control variable SIZE ranges from 12.671 to 19.317 with a mean of 15.516, revealing the diverse market power across the sample hospitality companies due to firm size effect. The fluctuation of firm age AGE is from 9 to 54 years with a mean of 33.831 and a standard deviation of 12.479. The dummy variable FAMILY has a mean of .692 and a standard deviation of .463, showing that the majority of the sample hospitality companies are controlled by family. 4. Methodology and results 4.1. Panel regression model The panel regression analysis is used to examine the determinants of the CP in the hospitality industry based on the value enhancement and agency cost theories. Panel data methodology can control for the heterogeneity of individual firms, reduce multicollinearity and estimation bias, and identify the time-vary structure between dependent and independent variables (Baltagi, 2005). The panel regression test equations used in the study are given as: Model I (the value enhancement theory): CS = ˛ + ˇ1 ADVR + ˇ2 PROT + ˇ3 LINT + ˇ4 RDR + ˇ5 CS(−1) + ˇ6 SIZE + ˇ7 AGE + ˇ8 FAMILY + ε.

(9)

Model II (the agency cost theory): CS = ˛ + ˇ1 DLEV + ˇ2 BS + ˇ3 INDBS + ˇ4 INSIDE + ˇ5 INSTIT + ˇ6 CS(−1) + ˇ7 SIZE + ˇ8 AGE + ˇ9 FAMILY + ε.

(10)

.000 .000 −12.540 .018 .000

.002 .019 5.146 .131 .529 × 10−4

When performing panel regression tests, we consider three estimation methods: pooled ordinary least square (OLS), fixed effects and random effects. Note that linear panel data models can be estimated using pooled OLS, the fixed effects method and the random effects method (Dimitrios, 2005). The pooled OLS method estimates the common constant for all cross-sections, implying that there are no differences between the estimated cross-sections. In the fixed effects method, the constant is treated as section-specific. The difference between the fixed effects method and random effects method is that the constants of the random effects method for each section are random parameters. Accordingly, the study uses F-test to determine whether the pooled OLS is more appropriate than the fixed effect method. The null hypothesis is that all the constants are the same and the common constant (pooled OLS) method is applicable. The null hypothesis can be rejected if F-statistic is larger than F-test critical value, implying that the fixed effects method is more appropriate than the pooled OLS method. Otherwise, the pooled OLS method is more appropriate than the fixed effects. Note that if the fixed effects method is more appropriate than the pooled OLS method, it is necessary to employ the Hausman (1978) test to examine whether fixed or random effects method should be used when performing panel regressions based on Eqs. (9) and (10). The F-test results (not reported here) indicate that panel regressions based on Eqs. (9) and (10) are significant at the 5% level, suggesting that the fixed effects method is more suitable for regression Eqs. (9) and (10). The Hausman (1978) test results (not reported here) also illustrate that panel regressions based on Eqs. (9) and (10) are significant at the 5% level. Hence, the fixed effects model is used to perform panel regressions for Eqs. (9) and (10).

4.2. Empirical results Before running the panel regression tests based on Eqs. (9) and (10), the study checks correlations between CP and philanthropy determinants, and correlations among all variables according to two theories. Among the four value-enhancing determinants, only PROT and LINT are significantly and positively correlated with CS over the full sample period (see panel A in Table 2). As for correlations among value enhancement determinants, ADVR is highly correlated with PROT and LINT, and PROT is highly correlated with LINT. Moreover, among five agency cost determinants, BS and INDBS

156

M.-H. Chen, C.-P. Lin / International Journal of Hospitality Management 48 (2015) 150–160

Table 2 Pearson correlations between corporate philanthropy and philanthropy determinants. Panel A: value enhancement determinants

CS

ADVR

PROT

LINT

RDR

CS(−1)

SIZE

AGE

FAMILY

CS ADVR PROT LINT RDR CS(−1) SIZE AGE FAMILY

1.000 −.027 (.71) .202 (.10)* .211 (.00)*** .023 (.75) .510 (.00)*** .011 (.88) .074 (.31) .105 (.11)

1.000 −.167 (.02)** .386 (.00)*** −.030 (.68) −.032 (.67) −.290 (.00)*** −.258 (.00)*** .199 (.01)***

1.000 −.271 (.00)*** −.069 (.34) .017 (.82) −.248 (.00)*** −.027 (.71) −.023 (.76)

1.000 .111 (.13) .135 (.06)* −.519 (.00)*** −.066 (.37) .303 (.00)***

1.000 .012 (.87) .151 (.04)** .103 (.16) −.020 (.79)

1.000 .020 (.78) .088 (.23) .101 (.12)

1.000 .079 (.28) −.152 (.04)**

1.000 −.175 (.02)**

1.000

Panel B: Agency cost determinants

CS

DLEV

BS

INDBS

INSITIT

INSIDE

CS(−1)

SIZE

AGE

FAMILY

CS DLEV BS INDBS INSITIT INSIDE CS(−1) SIZE AGE FAMILY

1.000 −.045 (.56) .272 (.00)*** −.240 (.10)* −.079 (.31) .108 (.16) .510 (.00)*** .033 (.67) .084 (.28) .105 (.11)

1.000 .283 (.00)*** −.060 (.44) −.072 (.36) .165 (.03)** −.061 (.44) .747 (.00)*** .056 (.47) .110 (.16)

1.000 −.044 (.57) −.176 (.02)** .177 (.02)** .276 (.00)*** .435 (.00)*** .075 (.34) .138 (.08)*

1.000 −.152 (.05)** .085 (.28) −.040 (.60) −.358 (.00)*** −.108 (.16) .266 (.00)***

1.000 .375 (.00)*** −.072 (.35) .017 (.83) −.196 (.01)** −.308 (.00)***

1.000 .124 (.11) .493 (.00)*** .211 (.01)*** −.359 (.00)***

1.000 .039 (.62) .094 (.23) .101 (.12)

1.000 .090 (.25) −.152 (.04)**

1.000 −.175 (.02)**

1.000

Note: figures in parentheses are p-values. * Significant at 10% level. ** Significant at 5% level. *** Significant at 1% level.

are significantly and positively correlated with CS (see panel B in Table 2). There are also high correlations among agency cost determinants. In addition, for the four control variables, CS(−1) is significantly and positively correlated with CS and some determinants (see panels A and B in Table 2). High correlations among value enhancement determinants and among agency cost determinants suggest the possibility of multicollinearity. Therefore, when running the panel regression test of the influences of all value enhancement (or agency cost) determinants on CP at the same time as in Eq. (9) (or Eq. (10)), the study cannot isolate the separate effects of individual value enhancement (or agency cost) determinants. In other words, it is difficult for this study to detect or isolate the true impact of each determinant within the model. The study hence performs several separate regression tests to avoid the possibility of multicollinearity. Table 3 reports the panel regression test results of Model I based on the value enhancement theory. Test results of Regression I in Table 3 reveal that among all four value enhancement determinants, only the positive coefficients of PROT and LINT are statistically significant at the 1% level. Results of panel regression analysis with each individual determinant and control variables further (Regressions II–V in Table 3) confirm that PROT and LINT are the only two value enhancement determinants that can significantly influence the CS. The explanatory power (adjusted R-square value) of Regression III and IV are .293 and .316, respectively. In contrast, results of the regression tests II and V point out that ADVR and RDR are insignificant in explaining the CP in the hospitality industry. Table 4 presents the panel regression test results for Model II based on the agency cost theory. Among the five agency cost determinants, only the coefficients of BS (0.107 × 10−3 ) and INDBS (−0.233 × 10−2 ) are significant at the 10% level (see Regressions I). Regression test results with each individual determinant and control variables (Regressions II–VI in Table 4) still show that only BS and INDBS are the agency cost determinants that are significant in explaining the CS. The explanatory power of Regression III and IV is .307 and .295, respectively. For the control variables (CS(−1), SIZE, AGE and FAMILY), regression test results find that CS(−1) is

significantly and positively related to CS and the impact of SIZE and AGE on CS is not statistically significant in general. To offer a robustness check for whether panel regression test results reported in Tables 3 and 4 are affected by a few influential observations of CS and hence can generate biased results, the study removes four influential observations and re-runs the regression tests. New panel regression test results of determinants of hospitality CP are summarized in Table 5. As shown in Table 5, the test results are not significantly different from those reported in Tables 3 and 4. Table 6 reports test results of regressions with only significant determinants, i.e. value enhancement determinants (LINT and PROT), agency cost determinants (BS and INDBS) and CS(−1). According to test results of regression I in Table 6, the significant effect of LINT, PROT, BS, INDBS and CS(−1) on CS remains when all five determinants are included in the regression test, suggesting the stable influence of the five factors on CS practices. Regression test results with each individual determinant show that each determinant can significantly affect CS individually. Among five determinants, BS has the highest value (.206) of the explanatory power, followed by LINT (.202), CS(−1) (.222), PROT (.059) and INDBS (.056).

5. Discussion and conclusion Given the limited data available, only a few empirical research studies have analyzed the motives of CP. Because CP data does not reveal whether altruistic consideration motivates CP (Fry et al., 1982), the aim of the study was to utilize the profit-maximizing and self-motivated managerial perquisite motives instead of altruism to examine the determinants of CP in the Taiwanese hospitality industry. This study is a starting point for hospitality readers and researchers to understand CP in the context of a hospitality industry. Empirical results of panel regression tests based on the value enhancement and agency cost theories can offer insights into the philanthropic behavior of the Taiwanese hospitality industry.

M.-H. Chen, C.-P. Lin / International Journal of Hospitality Management 48 (2015) 150–160

157

Table 3 Panel regression test results of determinants of hospitality corporate philanthropy based on the value enhancement theory. Independent variables

Regression I

Regression II

Regression III

Regression IV

Regression V

Regression VI

Constant ADVR PROT LINT RDR CS(−1) SIZE AGE Family F-statistic Adjusted R2

−.006 (.01)*** −.004 (.69) .933 × 10−4 (.01)*** .006 (.00)*** −1.141 (.72) .423 (.00)*** .199 × 10−3 (.11) .498 × 10−5 (.68) .131 × 10−3 (.70) 5.993*** .376

−.290 × 10−3 (.86) −.186 × 10−3 (.83)

−.904 × 10−3 (.56)

−.003 (.12)

−.313 × 10−3 (.84)

−.400 × 10−3 (.79)

1.139 (.72) .478 (.00)*** .165 × 10−5 (.98) .797 × 10−5 (.52) .423 × 10−3 (.21) 5.284*** .223

.479 (.00)*** .652 × 10−5 (.94) .861 × 10−5 (.48) .427 × 10−3 (.21) 5.282*** .217

Regression V

Regression VI

.612 × 10−4 (.06)* .003 (.01)***

.478 (.00)*** .140 × 10−5 (.98) .817 × 10−5 (.51) .438 × 10−3 (.20) 5.278*** .223

.460 (.00)*** .331 × 10−4 (.72) .854 × 10−5 (.46) .446 × 10−3 (.19) 5.669*** .293

.455 (.00)*** .137 × 10−3 (.18) .664 × 10−5 (.58) .201 × 10−3 (.56) 5.722*** .316

Note: figures in parentheses are p-values. * Significant at 10% level. ** Significant at 5% level. *** Significant at 1% level. Table 4 Panel regression test results of determinants of hospitality corporate philanthropy based on the agency cost theory. Independent variables Constant DLEV BS INDBS INSIDE INSTIT CS(−1) SIZE AGE FAMILY F-statistic Adjusted R2

Regression I

Regression II

.001 (.85) −.490 × 10−3 (.77) .107 × 10−3 (.07)* −.233 × 10−2 (.09)* −.404 (.77) .001 (.42) .451 (.00)*** −.938 × 10−4 (.68) .445 × 10−5 (.76) .373 × 10−3 (.40) 5.231*** .350

−.002 (.40) −.001 (.35)

.466 (.00)*** .113 × 10−3 (.44) .886 × 10−5 (.47) .567 × 10−3 (.13) 5.621*** .226

Regression III .001 (.83) .995 × 10−4 (.05)**

.448 (.00)*** .750 × 10−4 (.45) .671 × 10−5 (.58) .309 × 10−3 (.37) 5.935*** .307

Regression IV −4

−.510 × 10

(.97)

−.234 × 10−2 (.08)*

.475 (.00)*** .144 × 10−4 (.88) .813 × 10−5 (.51) .477 × 10−3 (.17) 5.715*** .295

−3

.543 × 10

(.75)

.180 × 10−3 (.87) .478 (.00)*** .724 × 10−5 (.94) .919 × 10−5 (.47) .449 × 10−3 (.22) 5.617*** .222

−.564 × 10−3 (.74)

.669 × 10−3 (.43) .496 (.00)*** .665 × 10−5 (.95) .768 × 10−5 (.58) .466 × 10−3 (.22) 5.622*** .239

Note: figures in parentheses are p-values. * Significant at 10% level. ** Significant at 5% level. *** Significant at 1% level. Table 5 A robustness check for panel regression test results of determinants of hospitality corporate philanthropy. Panel A: value enhancement

Regression I

Constant ADVR PROT LINT RDR – CS(−1) SIZE AGE FAMILY F-statistic Adjusted R2

−.004 (.06)* −.704 × 10−3 (.67) −.003 (.13) −.005 (.62) −4 *** −4 ** .902 × 10 (.03) .672 × 10 (.05) .005 (.01)*** .003 (.02)** −.390 (.91) .441 (.00)*** .201 × 10−3 (.11) .120 × 10−5 (.93) .502 × 10−3 (.23) 5.573*** .348

Regression II

.476 (.00)*** .190 × 10−4 (.84) .872 × 10−5 (.50) .538 × 10−3 (.20) 5.578*** .293

Regression III

.455 (.00)*** .139 × 10−3 (.19) .655 × 10−5 (.60) .188 × 10−3 (.61) 6.022*** .317

Panel B: agency cost

Regression I

Constant DLEV BS INDBS INSIDE INSTIT CS(−1) SIZE AGE FAMILY F-statistic Adjusted R2

.720 × 10−3 (.81) −.469 × 10−3 (.77) .108 × 10−3 (.07)* −.231 × 10−2 (.09)* −.377 (.79) .781 × 10−3 (.43) .451 (.00)*** −.936 × 10−4 (.69) .475 × 10−5 (.75) .368 × 10−3 (.41) 5.185*** .344

Regression II .312 × 10−3 (.84) .101 × 10−3 (.05)**

.447 (.00)*** −.745 × 10−4 (.47) .730 × 10−5 (.56) .308 × 10−3 (.39) 5.843*** .307

Regression III −.137 × 10−3 (.93)

−.229 × 10−2 (.08)*

.475 (.00)*** −.106 × 10−4 (.91) .882 × 10−5 (.49) .489 × 10−3 (.18) 5.622*** .294

Note: figures in parentheses are p-values. * Significant at 10% level. ** Significant at 5% level. *** Significant at 1% level.

Indeed, this study makes a crucial contribution by extending the investigation not only to the hospitality industry in the emerging country of Taiwan, but also to various academic researches related to CSR outside the hospitality industry and outside Taiwan. Although the study used the data of hospitality firms in Taiwan, the analysis of the CP (as a specific component of CSR) in the Taiwanese hospitality industry can be extended to other industries in different countries and/or to other hospitality segments, such as airline, gambling, hotel, restaurant and leisure and travel. Panel regression test results using 13 publicly traded hospitality companies from 1996 to 2011 indicated that several determinants

based on the two theories are significant in explaining CP in the hospitality industry. Specifically, two value enhancement determinants [labor intensity (LINT) and return on assets (PROT)], and two agency cost determinants [board size (BS) and independent directors-to-total directors (INDBS)] have been found to significantly influence the CP rate (CS). Moreover, the lagged philanthropy-to-sale ratio [CS(−1)] is significant in explaining the behavior of CP in the Taiwanese hospitality industry. The empirical findings and implications are presented and discussed as follows. First, this study found that both value enhancement and agency cost determinants are significant in explaining why Taiwanese

.479 (.00)*** 4.592*** .222 2.686** .055 3.435*** .206 4.370*** .242 Note: figures in parentheses are p-values. * Significant at 10% level. ** Significant at 5% level. *** Significant at 1% level.

3.360*** .202 4.417*** .245

2.720** .059

−.156 × 10−2 (.10)* .157 × 10−3 (.00)*** .156 × 10−3 (.00)*** −.171 × 10−2 (.07)* .566 × 10−4 (.09)*

Regression VII

.517 × 10−3 (.00)*** −.594 × 10−3 (.09)*

Regression VI Regression V

−.572 × 10−3 (.10)* .347 (.09)*

Regression IV

Constant LINT (value enhancement determinant) PROT (value enhancement determinant) BS (agency cost determinant) INDBS (agency cost determinant) CS(−1) F-statistic Adjusted R2

Regression III

−.968 × 10−4 (.64) .352 × 10−2 (.01)***

Regression II

−.001 (.10)* .004 (.00)*** .622 × 10−4 (.05)**

Regression I

−.001 (.01)*** .003 (.01)*** .617 × 10−4 (.05)** .104 × 10−3 (.03)** −.173 × 10−2 (.07)* .421 (.00)*** 6.597*** .447

Independent variables

Table 6 Results of panel regression tests with significant value enhancement and agency cost determinants.

.258 × 10−3 (.10)*

M.-H. Chen, C.-P. Lin / International Journal of Hospitality Management 48 (2015) 150–160

Regression VIII

158

hospitality firms contribute to philanthropy. This finding is different from the results of Brown et al. (2006), which indicated that agency cost determinants, not value enhancement determinants, play a major role in making philanthropy decisions. It is also inconsistent with the findings of Navarro (1988), which revealed that value enhancement determinants instead of managerial utilitymaximizing determinants are the main drivers of CP in the US firms. In addition, the explanatory power of two value enhancement determinants (LINT and PROT) on CP is .245 (Regression II in Table 6), which is very close to the corresponding value .242 of two agency cost determinants (BS and INDBS) (regression V in Table 6). Second, a significantly positive relationship emerged between CP and LINT. This finding suggests that the behavior of CP of publicly traded hospitality companies in Taiwan was driven partly by the desire to motivate the workforce. Furthermore, employee morale and commitment motivated by CP are similarly important to customer satisfaction because employees are in constant contact with the customers; thus, their attitudes would affect customers’ perceptions of service quality (Collier and Esteban, 2007). Accordingly, publicly traded Taiwanese hospitality companies with higher CP could improve their operating efficiency and reduce their operating cost by raising employee morale and productivity, which can then generate higher revenues and profits. Third, in addition to LINT, PROT is important in philanthropy decision. The study shows that the Taiwanese hospitality companies with a high level of profitability in terms of ROA contribute significantly more cash to philanthropy. This result confirms the findings of previous findings (Adams and Hardwick, 1988; Brammer and Millington, 2008; Roberts, 1992; Ullmann, 1985). Nonetheless, although profitability is one of two value enhancement determinants that can affect the hospitality industry’s CP in Taiwan, the explanatory power of PROT on CS is low at .059 compared to that of LINT on CS (.202). The size of the effect of LINT on hospitality CP is about 3.42 times of that of PROT. This suggests that between the two key value enhancement determinants, labor intensity rather than profitability is the major value enhancement determinant affecting the CP in the Taiwanese hospitality industry. Fourth, publicly traded hospitality companies with larger board size (BS) give significantly more cash to charity in Taiwan, showing that agency conflicts between the board and shareholders are significant determinants of the discretionary behavior of CP in the Taiwanese hospitality industry. These results imply that when engaging in CP, hospitality firms with a larger board size are more likely to increase conflicts between the board and shareholders. This suggests that shareholders of publicly traded hospitality firms in Taiwan should understand that to pursue their own interests, the board of directors tends to allocate a higher level of discretionary profit to CP. Shareholders of hospitality firms in Taiwan may need to evaluate the directors’ investment in CP more closely to verify that the investment does not exceed a reasonable or acceptable threshold, as the directors’ tendency to contribute to CP may come at their expense and eventually undermine financial performance. Fifth, not only a board’s size, but also its composition is relevant to the discretionary behavior of CP. The ratio of independent directors to total directors (INDBS) has also been found to be significantly and negatively associated with philanthropic contributions. This implies that independent directors of publicly traded hospitality companies in Taiwan do a better job ensuring that the managerial insiders’ interests are aligned with those of shareholders, and they could prevent the managerial insiders from advancing their own interests by CP at the expense of the shareholders. Thus, one way for the hospitality companies wishing to control their philanthropy contribution is to add more outside directors to the board. However, INDBS has a limited explanatory power (.056) on CS, which is relatively small compared to that of BS on CS (.206). This suggests that

M.-H. Chen, C.-P. Lin / International Journal of Hospitality Management 48 (2015) 150–160

between the two significant agency cost determinants, BS rather than INDBS is the major agency cost determinant of the CP in the Taiwanese hospitality industry. Sixth, this study reveals another new and interesting finding, implying that CS(−1) can also significantly explain the behavior of CP in the Taiwanese hospitality industry. In other words, hospitality firms with a history of philanthropy are more likely to continue contributing to philanthropy. Particularly, among five determinants that can affect CP, CS(−1) has the highest value (.222) of the explanatory power, followed by BS (.206), LINT (.202), PROT (.059), and INDBS (.056). This finding indicates that CS(−1) is very useful in explaining the behavior of CP in the Taiwanese hospitality industry. Indeed, financial or managerial stability of the companies could explain whether the behavior of corporate philanthropic activities is persistent (Roberts, 1992). Notably, changing philanthropy patterns of companies may signal financial or managerial instability of the companies, which may then lead to market-adverse expectations and reactions and damage the companies. Accordingly, the appropriate strategy for the companies wishing to avoid the risk of changing their philanthropy pattern is to continue their philanthropic practices. This strategy has also been implemented in the Taiwanese hospitality industry. Another possible explanation is that the persistence of CP may have significant implications for business management in the long term. As Simon (1995) noted, CP should be regarded as a component of long-term competitiveness rather than as a short-term image builder and sales generator. Thus, CP could be used as a form of long-turn business expense, i.e., a multi-period giving strategy. This study, however, had some limitations. First, while two value enhancement determinants (LINT and PROT), two agency cost determinants (BS and INDBS), and the CS(−1) can explain 22.2–25.5% of variations in the CP, all five factors together account for about 44.7% of variations in hospitality CP. Although the explanatory power of the determinants used in this study is higher compared to those found in previous studies, 55.3% of the variation in CP in the Taiwanese hospitality industry remains unexplained; thus, it can be affected by other factors. For example, both business condition and inbound tourism market growth are found to be significant factors insofar, as they both have a strong effect on corporate performance of publicly traded tourism firms in Taiwan (Chen, 2007, 2010). Further studies could further examine whether those two factors significantly explain CP decision in the Taiwanese hospitality industry. Second, while our results support the hypotheses that LINT, PROT, and BS are positively related to the philanthropy behavior of the Taiwanese hospitality industry while INDBS is negatively related to the CP decision of hospitality firms in Taiwan, the findings should be interpreted with caution. Since significant determinants of CP are likely to vary across sectors and countries, the results found in the Taiwanese hospitality industry may not be applied to the hospitality industry in different countries. For instance, Zhang et al. (2010) found no significant association between CP and PROT and Brown et al. (2006) failed to support a negative link between CP and INDBS. Future researchers can perform a comparative study using the data collected from hospitality-related firms in other developed and/or emerging countries to generate general conclusions. Finally, the empirical results of this study reveal that the value enhancement determinants (LINT and PROT) can explain why the hospitality corporations contribute to philanthropy. Since value enhancement theory proposes that CP can enhance corporate performance by creating value for shareholders, CP can be expected to affect the corporate performance of hospitality companies. Future studies could consider this assumption by investigating whether and to what extent CP affects the corporate performance of publicly traded hospitality companies.

159

Acknowledgments This research was conducted and completed when the first author spent his sabbatical leave at School of Hospitality Business Management, Washington State University. He is grateful to Washington State University for hosting him during his sabbatical leave. References Adams, M., Hardwick, P., 1988. An analysis of corporate donations: United Kingdom evidence. J. Manag. Stud. 35 (5), 641–654. Agrawal, A., Knoeber, C.R., 1996. Firm performance and mechanisms to control agency problems between managers and shareholders. J. Financ. Quant. Anal. 31 (3), 377–397. Amato, L.H., Amato, C., 2012. Retail philanthropy: firm size, industry, and business cycle. J. Bus. Ethics 107 (4), 435–447. Baltagi, B.H., 2005. Econometric Analysis of Panel Data, 3rd ed. John Wiley & Sons Ltd., West Sussex, England. Bartkus, B.R., Morris, S.A., Seifert, B., 2002. Governance and corporate philanthropy. Bus. Soc. 41 (3), 319–344. Brammer, S., Millington, A., 2004. The development of corporate charitable contributions in the UK: a stakeholder analysis. J. Manag. Stud. 41 (8), 1411–1434. Brammer, S., Millington, A., 2008. Does it pay to be different? An analysis of the relationship between corporate social and financial performance. Strateg. Manag. J. 29 (12), 1325–1343. Brown, W., Helland, E., Smith, K., 2006. Corporate philanthropic practices. J. Corp. Financ. 12 (5), 855–877. Chen, M.H., 2007. Interactions between business conditions and financial performance of tourism firms: evidence from China and Taiwan. Tour. Manag. 28 (1), 188–203. Chen, M.H., 2010. The economy, tourism growth and corporate performance in the Taiwanese hotel industry. Tour. Manag. 31 (5), 665–675. Chen, C.W., Yi, B., Lin, J.B., 2013. Media coverage, board structure and CEO compensation: evidence from Taiwan. J. Multinatl. Financ. Manag. 23 (5), 434–445. Chou, H.I., Chung, H., Yin, X., 2013. Attendance of board meetings and company performance: evidence from Taiwan. J. Bak. Financ. 37 (11), 4157–4171. Collier, J., Esteban, R., 2007. Corporate social responsibility and employee commitment. Bus. Ethics: Eur. Rev. 16 (1), 19–33. Cowan, A., Huang, C.H., Padmanabhan, P., Wang, C.H., 2013. The determinants of foreign giving: an exploratory empirical investigation of US manufacturing firms. Int. Bus. Rev. 22 (2), 407–420. Dimitrios, A., 2005. Applied Econometrics: A Modern Approach Using EViews and Microfit. Houndmills, Basingstoke, Hampshire, Palgrave Macmillan, New York. Du, X., Jian, W., Du, Y., Feng, W., Zeng, Q., 2013. Religion, the nature of ultimate owner, and corporate philanthropic giving: evidence from China. J. Bus. Ethics 81 (4), 1–22. Fama, E.F., Jensen, M.C., 1983. Separation of ownership and control. J. Law Econ. 26 (2), 301–325. Fry, L.W., Keim, G.D., Meiners, R.E., 1982. Corporate contributions: altruistic or for profit. Acad. Manag. J. 25 (1), 94–106. Godfrey, P.C., 2005. The relationship between corporate philanthropy and shareholder wealth: a risk management perspective. Acad. Manag. Rev. 30 (4), 777–798. Greening, D., Turban, D., 2000. Corporate social performance as a competitive advantage in attracting quality workforce. Bus. Soc. 39 (3), 254–280. Hart, O., 1993. An economist’s view of fiduciary duty. LSE Financial Markets Group Discussion Paper No. 157. London School of Economics, London, UK. Hausman, A., 1978. Specification tests in econometrics. Econometrica 46 (5), 1251–1271. Hsiao, C., 1986. Analysis of Panel Data. Cambridge University Press, Cambridge. Jensen, M., 1986. Agency costs of free cash flow, corporate finance, and takeovers. Am. Econ. Rev. 76 (2), 323–329. Jensen, M., 1993. The modern industrial revolution, exit and the failure of internal control systems. J. Financ. 48 (3), 831–880. Jensen, M.C., Meckling, W.H., 1976. Theory of the firm: managerial behavior, agency costs and ownership structure. J. Financ. Econ. 3 (3), 305–360. Lee, T.S., Yeh, Y.H., 2004. Corporate governance and financial distress: evidence from Taiwan. Corp. Gov.: Int. Rev. 12 (3), 378–388. Lev, B., Petrovits, C., Radhakrishnan, S., 2010. Is doing good for you? How corporate charitable contributions enhance revenue growth. Strateg. Manag. J. 31 (2), 182–200. Liu, Y., Ahlstrom, D., Yeh, K.S., 2006. The separation of ownership and management in Taiwan’s public companies: an empirical study. Int. Bus. Rev. 15 (4), 415–435. Navarro, P., 1988. Why do corporations give to charity? J. Bus. 61 (1), 65–93. Roberts, R.W., 1992. Determinates of corporate social responsibility disclosure: and application of stakeholder theory. Account. Org. Soc. 17 (6), 595–612. Ross, S.A., Westerfield, R.W., Jordan, B.D., 2007. Essentials of Corporate Finance, 5th ed. McGraw-Hill/Irwin, New York. Simon, F.L., 1995. Global corporate philanthropy: a strategic framework. Int. Mark. Rev. 12 (4), 20–37.

160

M.-H. Chen, C.-P. Lin / International Journal of Hospitality Management 48 (2015) 150–160

Turner, R., Miller, G., Gilbert, D., 2001. The role of UK charities and the tourism industry. Tour. Manag. 22 (5), 463–472. Ullmann, A.A., 1985. Data in search of a theory: a critical examination of the relationship among social performance, social disclosure, and economic performance. Acad. Manag. Rev. 10 (2), 540–557. Wang, J., Coffey, B.S., 1992. Board composition and corporate philanthropy. J. Bus. Ethics 11 (10), 771–778. Wang, H., Choi, J., Li, J., 2008. Too little or too much? Untangling the relationship between corporate philanthropy and firm financial performance. Org. Sci. 19 (1), 143–159.

Wright, M., Filatotchev, I., Hoskisson, R.E., Peng, M.W., 2005. Strategy research in emerging economies: challenging the conventional wisdom. J. Manag. Stud. 42 (1), 1–33. Yeh, Y.H., Lee, T.S., Woidtke, T., 2001. Family control and corporate governance: evidence from Taiwan. Int. Rev. Financ. 2 (1–2), 21–48. Young, M.N., Peng, M.W., Ahlstrom, D., Bruton, G.D., Jiang, Y., 2008. Corporate governance in emerging economies: a review of the principal–principal perspective. J. Manag. Stud. 45 (1), 196–220. Zhang, R., Zhu, J., Yue, H., Zhu, C., 2010. Corporate philanthropic giving, advertising intensity, and industry competition level. J. Bus. Ethics 94 (1), 39–52.