Corporate philanthropy in the US stock market: Evidence on corporate governance, value relevance and earnings manipulation

    Corporate Philanthropy in the US Stock Market: Evidence on Corporate Governance, Value Relevance and Earnings Manipulation George Emmanuel Iatridis PII: DOI: Reference:

S1057-5219(15)00041-1 doi: 10.1016/j.irfa.2015.03.004 FINANA 820

To appear in:

International Review of Financial Analysis

Received date: Revised date: Accepted date:

31 July 2014 10 February 2015 1 March 2015

Please cite this article as: Iatridis, G.E., Corporate Philanthropy in the US Stock Market: Evidence on Corporate Governance, Value Relevance and Earnings Manipulation, International Review of Financial Analysis (2015), doi: 10.1016/j.irfa.2015.03.004

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT TITLE: Corporate Philanthropy in the US Stock Market: Evidence on Corporate Governance, Value Relevance and Earnings Manipulation

RI P

T

AUTHOR: Dr George Emmanuel Iatridis Associate Professor of Accounting and Finance University of Thessaly, Department of Economics, Volos, Greece

SC

Visiting Professor at: EDHEC Business School, France, Grenoble Graduate School of Business, France, Universite Paris Dauphine, France, University of St Gallen, Switzerland, Management Center Innsbruck-MCI, Austria

MA NU

Editor, Accounting Subject International Review of Financial Analysis, Elsevier http://www.journals.elsevier.com/international-review-of-financial-analysis Research Fellow* Cass Business School, City University London, UK

PT

ED

* Research Fellowship outcome so far: Amiraslani, H., G. Iatridis and P. Pope (2013), “Accounting for Asset Impairment: A Test for IFRS Compliance across Europe”, A Research Report by the Centre for Financial Analysis and Reporting Research, Cass Business School, City University London, UK, ISBN: 978-0-9575905-1-9.

AC

CE

CORRESPONDENCE DETAILS: Dr George Emmanuel Iatridis 94 Vassani Street, Volos, 38 333, Greece E-mail: [email protected] Tel: +30 6973 963626

1

ACCEPTED MANUSCRIPT CORPORATE PHILANTHROPY IN THE US STOCK MARKET: EVIDENCE ON CORPORATE GOVERNANCE, VALUE RELEVANCE AND EARNINGS MANIPULATION

T

Abstract

RI P

This study examines the financial attributes of corporate philanthropy derived from the agency motives for corporate giving. Further, this study assesses the value

SC

relevance of corporate giving and investigates the impact of giving on investor perceptions and future profitability and growth. Also, it investigates the association

MA NU

between charitable spending and earnings manipulation. The findings indicate that the adoption of structured philanthropic initiatives and the use of in-kind contributions encourage corporate giving. The monitoring exercised by leverage and corporate governance affect corporate giving downwards. Firms that experience a management change are subject to more public scrutiny and tend to give more. Corporate giving is

ED

value relevant and is negatively related to analyst forecast error and positively to analyst coverage. Charitable firms tend to engage less in earnings manipulation. However, firms with significant growth options may direct slack resources to

PT

discretionary charitable causes. In-kind contributions are negatively related to

CE

managerial opportunism.

Keywords: Corporate

giving;

leverage;

litigation;

agency costs;

corporate

AC

governance; value relevance; earnings manipulation

JEL Classification: M41

1 Introduction Corporate philanthropy may be expressed as cash donations given to charities directly by the company or indirectly through a company-sponsored foundation, and/or as inkind gifts of a company‟s products, services, infrastructure, or know-how (Seifert et al, 2003). According to Giving USA 2012, charitable giving accounted for 2% of GDP in 2010. Giving USA 2012 reports that, in 2011, 73% of charitable giving came from individuals ($217.79 billion), 14% from foundations ($41.67 billion), 8% from bequests ($24.41 billion), and 5% from corporations ($14.55 billion). According to Giving in Numbers 2012, aggregate total giving has risen by 27% since 2009, the year

2

ACCEPTED MANUSCRIPT in which companies reported the most significant retreat in corporate giving. Since 2009, one-third of surveyed companies have increased their giving by 25% or more. According to Giving USA 2012, in 2011, the majority of charitable dollars

T

went to religion (32%), education (13%), and human services (12%). In 2011, total

RI P

(cash) giving as a percentage of pre-tax profits amounted to 0.95% (0.75%). Among the three giving types, non-cash was the most volatile, falling by 47% among companies giving less from 2010-2011 and increasing by 32% among companies

SC

giving more in that period. While cash and non-cash giving have both been on the rise, the proportion of cash to non-cash giving in aggregate has declined in recent

MA NU

years. Most companies cited general economic decline as the reason for a reduction in giving. However, in 2011, 45% of the companies that reported profit reductions increased their total giving. In 2011, 82% of corporate givers reported having a corporate foundation. According to the Center on Wealth and Philanthropy, foundation giving increased in 2011 to $41.67 billion, representing a 1.8% increase

ED

from 2010. According to The Chronicle of Philanthropy (July 14, 2013), the 10 companies that gave the most cash in 2012 are presented in Table 1.

PT

Mescon and Tilson (1987) argue that corporate giving can be considered as a form of corporate social behavior. Boatsman and Gupta (1996) argue that corporate

CE

giving is not just about maximizing company profits or improving certain financial figures, for example, through specific tax benefits, but it is about improving a company‟s social profile and picture. Corporate giving influences the social image of

AC

the company but is voluntary (Buchholtz et al, 1999). Corporate giving may be viewed positively by investors, market authorities, customers and the local community. Thus, although optional, it may be strongly recommended as a means of supporting and enhancing returns and future financial prospects (O‟Neill et al, 1989; Carroll, 1991). According to Seifert et al (2003), corporate philanthropy may be motivated by the following considerations. Companies may give strategically in order to reinforce their bottom line. They may also give because of social responsibility or legitimacy reasons, which would suggest that companies that perform well should support charitable activities (see Marquis et al, 2007). Another motivation for corporate philanthropy is to reinforce the CEO‟s profile, which may be attained easily by giving away shareholders‟ money. Companies may also engage in charitable practices due to other reasons, such as political costs, lobbying, or clientele considerations (Sanchez, 3

ACCEPTED MANUSCRIPT 2000). Gordon and Khumawala (1999) identify that other motives that encourage corporate giving include appreciation of the cause, company profitability, discretionary accruals, religion, and altruism.

T

Corporate philanthropy is significantly explained by the stakeholder theory in

RI P

the sense that it is a way for companies to display their social responsibility to the local community and satisfy stakeholders‟ interests (Clarkson, 1995; Berman et al, 1999). This study is mainly based on agency theory, which suggests that managers

SC

gain through corporate giving, and on value enhancement theory, which argues that corporate philanthropy increases shareholders‟ wealth. The motive to finance

MA NU

philanthropic causes in the effort to enhance a company‟s social responsibility profile may be regarded as an agency cost since the act of „good‟ by managers may create opportunity losses for shareholders. That is why powerful shareholders may generally be in favor of lower levels of corporate giving (Bartkus et al, 2002). Nevertheless, strategic philanthropy (Useem, 1988) would align the interests of those against with

ED

those in favor of corporate giving reducing any reservations or conflicts and leading to a new equilibrium (Buchholtz et al, 1999).

PT

Previous research on corporate philanthropy has concentrated on the association between corporate giving and taxes, company earnings, government

CE

incentives and market conditions (e.g. Boatsman and Gupta, 1996; Galaskiewicz, 1997; Seifert et al, 2003). This is the first study to link corporate giving to earnings manipulation and value relevance. It examines the financial attributes of corporate

AC

philanthropy derived from the agency motives for giving. It assesses the value relevance of corporate giving and investigates the impact of giving on investor perceptions and future profitability and growth. Also, it investigates the association between charitable spending and earnings manipulation. The US, which is the focus of this study, is shareholder/investor oriented, has strong investor protection mechanisms and effective corporate governance structures in place, and promotes adequate and relevant disclosures of corporate giving. The findings indicate that the adoption of structured philanthropic initiatives and the use of in-kind contributions encourage corporate giving. The monitoring exercised by leverage and corporate governance affect corporate giving downwards. Firms that experience a management change are subject to more public scrutiny and tend to give more. Corporate giving is value relevant and is negatively related to analyst forecast error and positively to analyst coverage. Charitable firms tend to 4

ACCEPTED MANUSCRIPT engage less in earnings manipulation. However, firms with significant growth options may direct slack resources to discretionary charitable causes. In-kind contributions are negatively related to managerial opportunism.

T

This study contributes by providing evidence that the public scrutiny to which

RI P

charitable spending is subjected has reduced the scope for earnings manipulation, even if charity givers display low liquidity and high leverage. This study implies that stricter charitable giving-related disclosures would be needed in order to mitigate the

SC

potential for earnings manipulation especially for firms with significant growth options that may be tempted to use financial surpluses for discretionary philanthropic

MA NU

causes. This study also distinguishes between in-kind and cash contributions, and indicates that, since in-kind contributions do not influence liquidity levels and thus do not generate opportunistic behaviors, they may be used as a monitoring tool to reduce the levels of and the potential for managerial opportunism. The remaining sections of the study are as follows. Section 2 presents

ED

background considerations. Section 3 shows the research hypotheses. Section 4 presents the datasets and limitations of the study. Section 5 discusses the empirical

PT

findings, and Section 6 presents the conclusions of the study.

CE

2 Background Considerations 2.1 Attributes of and Motives for Corporate Giving Higher liquidity would be positively linked to corporate giving (Pallot, 1990; Parsons

AC

and Trussel, 2008). Higher levels of debt would reduce free cash flows and limit the potential for wasting resources or discretionarily misdirecting funds (Harvey et al, 2004; Brown et al, 2006; Zhang et al, 2010). Thus, high leverage and strict debt covenants are likely to reduce corporate giving (Brown et al, 2006; Zhang et al, 2010). Higher debt would signify that lenders exercise greater monitoring, and thus corporate giving would be more credible. Therefore, corporate giving would be expected to be of higher quality and to a greater extent targeted to objective and real causes. Higher adequacy of equity (net assets) would reflect companies‟ ongoing ability to make donations (Trussel and Greenlee, 2004). High growth prospects would show that companies possess certain competitive advantages and managerial skills, which may lead to positive market values and stock returns and encourage them to give to charity. It could also be argued of course that growth companies may give less 5

ACCEPTED MANUSCRIPT in order to use excessive cash for their growth potential, implying that corporate giving is a managerial decision that is highly discretionary. Corporate giving would also be influenced by market structure and market conditions. For example,

T

companies that operate in a highly competitive market might earn lower or zero

RI P

abnormal returns, which may limit corporate giving.

Companies with diverse sources of revenue would possess a more stable income stream, which would maintain and reinforce their corporate giving in the

SC

long-run (Greenlee and Trussel, 2000; Parsons and Trussel, 2008; Trussel and Parsons, 2008). Galaskiewicz (1997) found a positive relationship between corporate

MA NU

giving and return on sales, return on assets and return on equity. Also, income tax considerations are very important when deciding the amount and the target of corporate giving (Boatsman and Gupta, 1996). The actual contribution should be higher than the cost incurred to the donor when contributions are tax-deductible. Aggarwal et al (2012) argue that companies set managerial goals, including

ED

community welfare and development, environmental protection and non-profit causes, which promote shareholder value maximization from a broader perspective. Thus,

PT

large and highly visible companies, such as telecommunications and media, may be inclined to give back to the community in the form of public donations in order to

CE

obtain a favorable treatment from market participants and authorities (Wood and Jones, 1995; Galaskiewicz, 1997; Buchholtz et al, 1999; Amato and Amato, 2007). Similar considerations would hold for companies that operate in environmentally

AC

sensitive industries, such as chemicals, plastics, oil and gas, as well as for companies that are in highly regulated industries, such as banks and financial services. Companies may be motivated to create charitable foundations, which may shield them from business cycle fluctuations and smooth earnings by differently influencing the transfers to the foundation in good and bad economic years (Brown et al, 2006). Corporate giving would tend to be positively related to board size. Larger boards would get involved with more non-financial objectives that go beyond the narrowly defined financial performance maximization targets (Aggarwal et al, 2012). In large boards, more directors may be inclined to pursue their philanthropical objectives leading to higher corporate giving, especially if charitable giving is perceived to improve their public image and reputation. Boards with more independent and outside directors would tend to apply more intensive oversight (Bhagat and Jefferis, 2002; Brammer and Millington 2005; Brown et al, 2006). 6

ACCEPTED MANUSCRIPT

2.2 Agency Theory Considerations McGuire et al (1988) and Waddock and Graves (1997) have reported that the intensity

T

of corporate giving relies on the accumulation of resources (see also Wood and Jones,

RI P

1995; Rowley and Berman, 2000). These resources may be scarce, difficult to imitate or to substitute (Amit and Schoemaker, 1993). Seifert et al (2003) argue that since corporate giving is considered as discretionary, it would also fluctuate with the level

SC

of discretionary accruals. Seifert et al (2003) also indicate that free cash flows may be used to interpret the behavior of discretionary accruals, and therefore, high free cash

MA NU

flows would be positively related to high corporate giving (see Buchholtz et al, 1999; Wang et al, 2008).1 This is in line with Jensen (1986), who notes that free cash flows are cash flows that are left over after financing projects with positive net present values. It follows that managers‟ interest and discretion on these cash flows is significant.

ED

Agency theory suggests that corporate giving may be important when it serves managerial objectives, such as to maximize shareholder wealth or manager

PT

compensation. Agency theory argues that information asymmetry may motivate managers to make earnings-increasing decisions that reinforce their personal wealth

CE

even if they are at the expense of shareholders‟ long-term financial interests (Jensen and Meckling, 1976; Healy and Palepu, 2001). So, managers may act opportunistically by using shareholder money or giving to targeted beneficiaries at

AC

their discretion in order to achieve self-defined goals or obtain personal influence (see also Galaskiewicz, 1997; Bartkus et al, 2002; Brown et al, 2006). Haley (1991) indicates that corporate giving may be regarded as a form of social exchange or discretionary reserve for managers, which may lead to further managerial opportunism. The fact that managers may give to social causes instead of investing these free cash flows to profitable investment projects or returning them to shareholders would give rise to agency costs (Jensen, 1986). Shareholders may prefer receiving company year earnings residuals instead of distributing them in the form of donations. Only if the use of free cash flows for corporate giving and contributions surpasses the related opportunity costs and gives rise to strategic benefits will shareholders‟ returns be reinforced. In efficient markets, the conflicts between managers and shareholders would be expected to establish effective corporate 1

Mitchell et al (1997), Seifert et al (2003) and Wang et al (2008) argue that free cash flows capture a company‟s slack resources better than profits.

7

ACCEPTED MANUSCRIPT governance mechanisms and monitoring devices of manager actions in order to reduce the possibilities for manipulation or misuse of company resources or shareholder residuals. Intensive monitoring would tend to limit the scope for

T

managing accruals (Jensen, 1986). Transparent and prudent corporate giving would

RI P

tend to reduce agency costs and the need for stricter monitoring mechanisms, and would also benefit company shareholders.

Strategic philanthropy may align the interests of managers with those of

SC

shareholders especially when corporate giving is aimed at obtaining a competitive advantage, enhancing the company‟s brand name and growth and improving

MA NU

corporate profits (Fombrun and Shanley, 1990; Brammer and Millington, 2005). Corporate giving may also take non-cash forms, such as in-kind gifts (Seifert et al, 2003). Cash contributions would be closely linked to discretionary accruals and potentially to opportunistic manipulations. Agency theory would suggest that shareholders would tend to be more interested in cash contributions as opposed to in-

ED

kind gifts. On the other hand, although a company could earn revenues by selling an in-kind gift, managers could select to donate in-kind gifts, which are of lesser

PT

operational value for the company or which entail relatively less significant outflows. Thus, strategic philanthropy might involve giving targeted in-kind gifts, such as

CE

tangible fixed assets that are close to the end of their useful economic life or obsolete inventory, or low-cost goods/services to socially sensitive groups of people for free, in order to cut costs, reinforce earnings and still influence stakeholders‟ perceptions in a

AC

manner similar to cash contributions. Also, a company could obtain similar social benefits by improving their public picture through the reduction of employee layoffs, investment in local community infrastructure, etc (Seifert et al, 2003). Hence, corporate giving may be viewed as a source of value and risk mitigation for shareholders (Brown et al, 2006; Johanson and Vahlne, 2006).

3 Research Hypotheses 3.1 Financial Attributes of Corporate Giving The presence of high operating margins would reflect a higher potential surplus and would therefore suggest that companies would be more likely to engage in charitable activities (Greenlee and Trussel, 2000). The existence of effective corporate governance mechanisms would ensure that the funds given to charitable objectives are based on transparent and beyond dispute decisions and that they do not serve 8

ACCEPTED MANUSCRIPT managerial opportunism or other self-defined purposes. Here, it is hypothesized that increased outside monitoring would lead to less charitable giving but of higher quality (Seifert et al, 2003). More intensive monitoring would also be expected in the case of

T

concentrated ownership, where shareholders are more actively involved in managers‟

RI P

oversight (Brown et al, 2006).

Companies that face litigation risks, rate or entry regulation are likely to undertake charitable initiatives in order to positively influence regulators‟ perceptions

SC

and attitude (Brown et al, 2006). Brown et al (2006) suggest that such firms tend to display higher R&D expenditure and advertizing and corporate giving. Corporate

MA NU

philanthropy is essentially a form of advertizing. On the other hand, high R&D expenditure may be related to intangible assets that are more likely to be affected by lawsuits or government regulation, thereby motivating companies to increase corporate giving in order to improve their public image and influence potential court decisions and regulators (Brown et al, 2006). The hypothesis that is tested is presented

Corporate giving is likely to display a positive association with in-kind

PT

H1

ED

below.

contributions, size, liquidity, R&D expenditure and the possibility of litigation,

CE

and a negative association with leverage, agency costs, growth and corporate governance.

AC

CGi,t = a0 + a1 PHILi,t + a2 INCi,t + a3 lnFREi,t-1 + a4 ROAi,t-1 + a5 GEARi,t-1 + a6 lnAi,t-1 + a7 TQi,t-1 + a8 CFLOWi,t-1 + a9 MBVi,t-1 + a10 ADVERTi,t-1 + a11 R&Di,t-1 + a12 INDi,t-1 + a13 IDBi,t-1 + a14 NEXECi,t-1 + a15 MCi,t-1 + a16 REMi,t-1 + a17 BIGAUi,t-1 + a18 NAi,t-1 + a19 SOi,t-1 + a20 CLSi,t-1 + a21 LITi,t-1 + e where CGi,t PHILi,t INCi,t lnFREi,t-1 ROAi,t-1 GEARi,t-1 lnAi,t-1 TQi,t-1

(1)

is a proxy for corporate giving and is annual corporate giving scaled by total assets. is a dummy variable and proxies for philanthropy. PHILi,t = 1 for firms that adopt structured philanthropic initiatives and PHILi,t = 0 otherwise. is a firm‟s in-kind contributions as a proportion to total corporate giving. is the natural logarithm of fundraising expenses. is return on assets and is income before extraordinary items scaled by total assets. is debt scaled by total assets. is the natural logarithm of total assets. is Tobin’s Q and is equal to market value of common equity plus book value of preferred stock, plus book value of long9

ACCEPTED MANUSCRIPT

REMi,t-1 BIGAUi,t-1 NAi,t-1 SOi,t-1

T

ED

CLSi,t-1

RI P

NEXECi,t-1 MCi,t-1

SC

MBVi,t-1 ADVERTi,t-1 R&Di,t-1 INDi,t-1 IDBi,t-1

MA NU

CFLOWi,t-1

term debt and current liabilities, scaled by book value of total assets. is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. is market to book value of equity. is advertizing expenses scaled by total sales. is research and development expenditure scaled by total sales. is a dummy variable and proxies for industry classification. is the percentage of independent directors sitting on the board of directors. is the ratio of non-executive to executive directors. MCi,t-1 = 1 when changes in the management have occurred and MCi,t-1 = 0 otherwise. is performance-related bonus. BIGAUi,t-1 = 1 for firms that are audited by a big 4 auditor and BIGAUi,t-1 = 0 otherwise. is a proxy for stability and is net assets scaled by total revenue (see Trussel and Parsons, 2008). is the proportion of ordinary shares held by shareholders, who own 5% or more of the company (see Brown et al, 2006). is the percentage of shares closely held by strategic investors, such as families, trusts, foundations and institutional investors. is the probability of litigation and is estimated using the following probit model (see Rogers and Stocken, 2005, pp. 1245, 1256-1257). is the error term.

PT

LITi,t-1 e

LS = 1 if a securities class action lawsuit was recorded by the Stanford Law School‟s Securities Class Action Clearinghouse during the period under investigation and LS = 0 otherwise. ND is the normal cumulative distribution function. MVE is the natural logarithm of market value of equity. ST is trading volume scaled by market value of outstanding equity. BETA is the slope coefficient and is estimated from regressing returns on the CRSP equal-weighted index. R is stock return. RV is the standard deviation of stock returns. SK is the skewness of stock returns. MR is the minimum of stock returns. HRI is a dummy variable and proxies for high risk industries. HRI includes Biotechnology (SIC 2833 to 2836), Computer Hardware (SIC 3570 to 3577), Electronics (SIC 3600 to 3674), Retailing (SIC 5200 to 5961) and Computer Software (SIC 7371 to 7379). e is the error term.

AC

LS

CE

Probit (LS = 1) = ND (α0 + α1 MVE + α2 ST + α3 BETA + α4 R + α5 RV + α6 SK + α7 MR + Σ HRI + e) (2)

10

ACCEPTED MANUSCRIPT It is noted that contribution variables, such as CGi,t, INCi,t and PHILi,t, are measured in period t0 as significant changes in company funding or charity programs would become promptly known to interested parties and market participants

T

(Tinkelman, 1999; Trussel and Parsons, 2008). Since corporate philanthropy is

RI P

closely dependent upon resources that are already available (Waddock and Graves, 1997; Buchholtz et al, 1999), this study uses lagged financial performance variables. In support of this view, McGuire et al (1988) and Berman et al (1999) have found

SC

significant association between charitable giving and lagged financial performance. The use of lagged variables is important because companies that engage in

MA NU

philanthropic activities would not be aware of their current financial performance when making decisions about current year‟s charitable giving levels (Trussel and Parsons, 2008).

3.2 Corporate Giving, Investor Perceptions and Value Relevance: Is Corporate

ED

Giving Worthwhile?

In the Holy Bible it is written „Give, and it will be given to you‟ (Luke 6: 38). Value

PT

enhancement theory suggests that corporate giving reinforces the financial and nonfinancial worth of a company (Brown et al, 2006). Reputation may be achieved by

CE

corporate philanthropy, which would subsequently generate tangible future benefits and profits (Roberts and Dowling, 2002; Brammer and Millington, 2005). Companies that give are likely to attract investors‟ favor, which might also take the form of easier

AC

access to capital markets and lower cost of capital. This would positively influence companies‟ future financial prospects and firm value (Plumlee et al, 2010; Dhaliwal et al, 2011). Orlitzky et al (2003) have found a positive association between corporate giving and subsequent financial performance. In contrast, the absence of corporate giving may give negative signals of selfishness or lack of cash flows and may result in limited abnormal returns for shareholders. Givers would be expected to display higher stock returns through their sharing of the local community‟s financial or other burdens and through market participants‟ appreciation and approval. The hypothesis that is tested is as follows.

H2

Corporate giving is value relevant and is likely to positively influence investor perceptions and future profitability and growth.

11

ACCEPTED MANUSCRIPT To test the value relevance part of this hypothesis, this study uses the Ohlson (1995) valuation equation as adopted by Clarkson et al (2013, p. 10). Market and contribution variables, such as Pi,t, CGi,t, PHILi,t and INCi,t, are measured in period t0

T

as information about company funding or charity programs would become quickly

RI P

known. The use of lagged financial variables is vital because market participants would not have access to such financial information in the year that corporate giving

SC

occurred (Trussel and Parsons, 2008).

is the stock price at the end of the year. is book value of equity. is abnormal earnings per share. AEPSi,t-1 is earnings per share at the end of the year, minus the cost of equity capital multiplied by book value of equity at the beginning of the year. To estimate the cost of equity capital (CEQ), this study employs the price to earnings growth ratio based on Easton (2004) as adopted by Clarkson et al (2013, p. 13):

ED

where Pi,t BVi,t-1 AEPSi,t-1

MA NU

Pi,t = a0 + a1 BVi,t-1 + a2 AEPSi,t-1 + a3 CGi,t + a4 PHILi,t + a5 INCi,t + a6 CFLOWi,t-1 + a7 CFLOWi,t-1 x CGi,t + a8 CFLOWi,t-1 x PHILi,t + e (3)

PT

CEQ = √ [(f2 – f1) / p0 ]

AC

CGi,t

CE

where

PHILi,t INCi,t

CFLOWi,t-1 e

f2

f1

p0

(4)

is a two-year median forecast of earnings per share. is a one-year median forecast of earnings per share. is share price.

is a proxy for corporate giving and is annual corporate giving scaled by total assets. is a dummy variable and proxies for philanthropy. PHILi,t = 1 for firms that adopt structured philanthropic initiatives and PHILi,t = 0 otherwise. is a firm‟s in-kind contributions as a proportion to total corporate giving. is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. is the error term.

To test value relevance, this study also uses the model below, which is based on Cormier and Magnan (2007, p. 617). MBVi,t = a0 + a1 1/BVi,t-1 + a2 EAR/BVi,t-1 + a3 CGi,t + a4 PHILi,t + a5 INCi,t + a6 EAR/BVi,t-1 x CGi,t + a7 EAR/BVi,t-1 x PHILi,t + a8 CFLOWi,t-1 + a9 CFLOWi,t-1 x CGi,t + a10 CFLOWi,t-1 x PHILi,t + e 12

(5)

ACCEPTED MANUSCRIPT

INCi,t

MA NU

CFLOWi,t-1

T

PHILi,t

RI P

CGi,t

is market to book value of equity. is book value of equity. is net income before extraordinary items scaled by book value of equity. is a proxy for corporate giving and is annual corporate giving scaled by total assets. is a dummy variable and proxies for philanthropy. PHILi,t = 1 for firms that adopt structured philanthropic initiatives and PHILi,t = 0 otherwise. is a firm‟s in-kind contributions as a proportion to total corporate giving. is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. is the error term.

SC

where MBVi,t BVi,t-1 EAR/BVi,t-1

e

Further, this section examines the empirical association between corporate giving and analyst earnings forecast accuracy, dispersion and revision volatility, returns, investor perceptions and future profitability, growth and foreign sales.

ED

Because, frequently, a company gives cash to its foundation in a specific year and, in turn, the foundation makes donations to the rightful recipients over a number of

PT

subsequent years, this study uses leading variables as well (see Seifert et al, 2003).

CE

The regression equation is presented below.

AC

CGi,t = a0 + a1 FE + a2 FD + a3 logAF + a4 RO + a5 ST + a6 CTV + a7 BA + a8 Ri,t + a9 Posi,t + a10 Ri,t x Posi,t + a11 Ri,t+1 + a12 Posi,t+1 + a13 Ri,t+1 x Posi,t+1 + a14 RV + a15 JF + a16 ROAi,t+1 + a17 MBVi,t+1 + a18 INTERi,t+1 + e (6) where CGi,t FE

FD

logAF RO

ST CTV

is a proxy for corporate giving and is annual corporate giving scaled by total assets. is analyst forecast error, i.e. mean minus actual forecast, scaled by absolute actual earnings (see Cotter et al, 2012; Horton et al, 2013). is forecast dispersion and is the standard deviation of analyst forecasts scaled by absolute actual earnings (see Cotter et al, 2012; Horton et al, 2013). is the log of one plus the number of analysts covering the firm (see Muller et al, 2008). is revision optimism and is the difference between upward revisions and downward revisions scaled by the number of estimates (see Horton et al, 2008). is trading volume scaled by market value of outstanding equity (see Daske et al, 2013). proxies for stock illiquidity and is the price change in trading volume (see Amihud, 2002; Daske et al, 2013).

13

ACCEPTED MANUSCRIPT

JF = (f2-fu) / (f+u)2, if f>u,

(7)

JF = (fu-u2) / (f+u)2, if u>f

(8)

f u

PT

is return on assets and is income before extraordinary items scaled by total assets. is market to book value of equity. is foreign sales scaled by total sales. is the error term.

CE

MBVi,t+1 INTERi,t+1 e

is the number of favorable corporate giving articles obtained from Factiva. is the number of unfavorable corporate giving articles obtained from Factiva.

ED

where

ROAi,t+1

MA NU

SC

RI P

Ri,t, Ri,t+1 Posi,t, Posi,t+1 RV JF

is the bid-ask spread, which proxies for information asymmetry (see Healy et al, 1999; Leuz and Verrecchia, 2000; Daske et al, 2013). are stock returns. Pos = 1 for positive return values and Pos = 0 otherwise. is the standard deviation of returns. is the Janis-Fadner coefficient (see Janis and Fadner, 1965; Aerts and Cormier, 2009). JF measures the favorability of media coverage regarding companies‟ corporate giving. JF takes values from –1 to +1. –1 reflects unfavorable perceptions. +1 reflects favorable perceptions. 0 would imply neutral perceptions. This study has used the Factiva database to search for news stories. JF is computed as follows (see Janis and Fadner, 1965; Clarkson et al, 2013, p. 15):

T

BA

AC

3.3 Corporate Giving and Discretionary Accruals Agency theory shows that, under corporate giving, companies donate funds that could otherwise be used in positive NPV projects. If there are no significant investment opportunities, companies should return the money to its proprietary owners, i.e. the shareholders (Jensen, 1986). Thus, as an alternative outlet, corporate giving would tend to increase as free cash flows increase (Seifert et al, 2003). Hence, it is hypothesized that companies that engage in charitable activities would have available resources to support their giving decision and would not undertake earnings manipulation practices. Of course, agency theory suggests that managers may act opportunistically and neglect their shareholder wealth maximization role in order to promote their selfdefined interests. With respect to corporate philanthropy, their self-defined interests may be expressed in terms of directing slack resources to specific charitable causes in 14

ACCEPTED MANUSCRIPT order to improve their prestige or obtain recognition from local community (Galaskiewicz, 1985; Seifert et al, 2003). However, the presence of blockholders as well as of non-executive and independent directors on company boards or being

T

audited by a big auditor would exercise an effective oversight role, which would

RI P

reduce the scope for earnings manipulation.

In line with the arguments discussed in Section 3.1 and Peasnell et al (2005), this study uses a mixture of current and lagged independent variables and would

SC

expect that engaging in earnings management (if at all) in the previous year would affect current year‟s cash flows and earnings and thus current year‟s corporate giving

H3

MA NU

decisions. The hypothesis is presented below.

Firms that engage in charitable giving are likely to exhibit lower discretionary accruals.

PT

ED

AAi,t-1 = a0 + a1 CGi,t + a2 PHILi,t + a3 INCi,t + a4 lnFREi,t-1 + a5 BELi,t-1 + a6 HIGHi,t-1 + a7 BELi,t-1 x CGi,t + a8 HIGHi,t-1 x CGi,t + a9 ICFi,t-1 + a10 lnAi,t-1 + a11 CFLOWi,t-1 + a12 MBVi,t-1 + a13 CGi,t x lnAi,t-1 + a14 CGi,t x CFLOWi,t-1 + a15 CGi,t x MBVi,t-1 + a16 IDBi,t-1 + a17 NEXECi,t-1 + a18 BIGAUi,t-1 + a19 CLSi,t-1 + e (9) is abnormal accruals and obtained as follows (see Dechow et al, 1995; Peasnell et al, 2005, pp. 1320-1321).

CE

where AAi,t-1

AC

AAi,t-1 = WCi,t-1 / TAi,t-2 – [ where

WCi,t-1 TAi,t-2 ΔREVi,t-1 ΔRECi,t-1

and

(1 / TAi,t-2) +

(ΔREVi,t-1 - ΔRECi,t-1) / TAi,t-2] (10)

is working capital computed as the change in con-cash current assets less the change in current liabilities. is total assets. is the change in revenues. is the change in receivables. According to Dechow et al (1995), ΔREVi,t-1 - ΔRECi,t-1 accounts for credit sales manipulations. are the OLS regression estimates of α0 and α1 respectively obtained from equation (11) based on the modified Jones model.

WCi,t-1 / TAi,t-2 = α0 (1 / TAi,t-2) + α1 (ΔREVi,t-1 - ΔRECi,t-1) / TAi,t-2 + e where

WCi,t-1 TAi,t-2 ΔREVi,t-1

(11)

is working capital computed as the change in con-cash current assets less the change in current liabilities. is total assets. is the change in revenues. 15

ACCEPTED MANUSCRIPT ΔRECi,t-1 e

HIGHi,t-1

ICFi,t-1

T

RI P

CE

lnAi,t-1 CFLOWi,t-1

SC

lnFREi,t-1 BELi,t-1

MA NU

INCi,t

ED

PHILi,t

is a proxy for corporate giving and is annual corporate giving scaled by total assets. is a dummy variable and proxies for philanthropy. PHILi,t = 1 for firms that adopt structured philanthropic initiatives and PHILi,t = 0 otherwise. is a firm‟s in-kind contributions as a proportion to total corporate giving. is the natural logarithm of fundraising expenses. BELi,t-1 = 1 if earnings before manipulation, i.e. reported earnings minus abnormal accruals, scaled by total assets, is less than zero or less than lagged reported earnings and 0 otherwise (see Peasnell et al, 2005, p. 1322). HIGHi,t-1 = 1 if earnings before manipulation, i.e. reported earnings minus abnormal accruals, scaled by total assets, is more than zero or more than lagged reported earnings and 0 otherwise (see Peasnell et al, 2005, p. 1322). is the standard deviation of operating income scaled by total assets divided by the standard deviation of operating cash flow scaled by total assets (see Leuz et al, 2003; Ding et al, 2007, p. 25). is the natural logarithm of total assets. is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. is market to book value of equity. is the percentage of independent directors sitting on the board of directors. is the ratio of non-executive to executive directors. BIGAUi,t-1 = 1 for firms that are audited by a big 4 auditor and BIGAUi,t-1 = 0 otherwise. is the percentage of shares closely held by strategic investors, such as families, trusts, foundations and institutional investors. is the error term.

PT

CGi,t

is the change in receivables. According to Dechow et al (1995), ΔREVi,t-1 - ΔRECi,t-1 accounts for credit sales manipulations. is the error term.

AC

MBVi,t-1 IDBi,t-1

NEXECi,t-1 BIGAUi,t-1

CLSi,t-1 e

4 Datasets and Limitations 4.1 Datasets The sample consists of companies listed on the New York Stock Exchange (NYSE). Banks, insurance, pension and brokerage firms have been excluded, as their accounting methods are not always comparable with those of industrial firms, thus limiting the sample to 2,152 companies. To obtain charitable giving data, this study used data provided in the Corporate Giving Directory, Giving USA, and the National Directory of Corporate Giving. The sample of 2,152 companies was matched with 16

ACCEPTED MANUSCRIPT charitable data from the giving directories presented above, with financial data from DataStream, with corporate governance data from BoardEx and SEC filings, and with financial analyst data from Thomson One Banker. Due to data constraints, and

T

particularly because not all companies spend on charity, the final sample consists of

RI P

687 companies. The period of investigation is the five-year period from 2008 to 2012. This study has accounted for heteroscedasticity, autocorrelation, departure from normality and multicollinearity, where appropriate. The tests

SC

that have been performed to check the OLS assumptions are the White test and the Autoregressive Conditional Heteroscedasticity (ARCH) test for

MA NU

heteroscedasticity; the Durbin-Watson test and the Breusch-Godfrey test for autocorrelation; the Jarque-Bera test for the departure from normality of residuals; and the correlation coefficients among the test variables for multicollinearity.

ED

4.2 Limitations

This study is limited in the following respects. The literature argues that there is a

PT

certain lack of theorization or conceptualization on the causal relationships between corporate giving and financial performance, reputation, and the other issues explored

CE

in this study (see Margolis and Walsh, 2003; Godfrey, 2005; Vogel, 2005). Further, there are measurement issues, which may make comparisons less effective. For example, when measuring financial performance, accounting or market measures may

AC

be used. However, certain companies prefer in-kind charitable giving, which may not be reported, and thus not captured (Vaidyanathan, 2008). On the other hand, when measuring social performance, Vaidyanathan (2008, p. 39) notes that different tools may be used, such as environmental awards, mutual funds screens, ratings obtained from the Council on Economic Priorities, the Kinder, Lydenberg and Domini ratings, Fortune reputation survey ratings, etc. However, these indices may be linked to different limitations and problems.

5 Empirical Findings 5.1 Descriptive Statistics Table 2, Panel A, presents the descriptive statistics for the main variables used in the empirical analysis. The mean annual contribution (CGi,t) is $822 per million dollars of assets. The mean of fundraising expenses (lnFREi,t-1) is 7.5. The mean return on assets 17

ACCEPTED MANUSCRIPT (ROAi,t-1) is 7%. The mean debt to assets (GEARi,t-1) is 53%. The mean obtained for size (lnAi,t-1) is 14. The mean cash flow ratio (CFLOWi,t-1) is 6% and the mean market to book value of equity (MBVi,t-1) is 3.3. The mean advertizing expenses (ADVERTi,t-1)

T

and the mean research and development expenditure (R&Di,t-1) amount to 3.2% and

RI P

4.6% respectively. On average, 44% of the directors sitting on the board are independent (IDBi,t-1). The typical sample firm has a mean percentage of shares closely held by strategic investors (CLSi,t-1) of 24%. The mean forecast error (FE)

SC

amounts only to 2% reflecting higher forecast accuracy. The trading volume (ST) displays a mean of 0.52. The mean bid-ask spread (BA) is only 0.02. The mean

MA NU

reported for abnormal accruals (AAi,t-1) is also small and amounts to 0.09. Panel B shows the descriptive statistics per sample year and indicates that the mean corporate giving per million $ assets increases from 809 in 2008 to 833 in 2012. Most of the giving is concentrated in the 75th quartile for every sample year, although it presents some degree of fluctuation from year to year. The greatest value, 1,086, is

ED

reported in 2012. The highest median, 372, is also reported in 2012. With regard to

PT

the 25th quartile, the greatest value, 172, is reported in 2010.

5.2 Financial Attributes of Corporate Giving

CE

The findings presented in Table 3 give support to H1 and indicate that corporate giving displays a significantly positive association with in-kind contributions, size, liquidity, R&D expenditure and the possibility of litigation. In contrast, corporate

AC

giving is negatively associated with leverage, agency costs, growth and corporate governance. For robustness purposes, Appendix 1 presents the findings without structured philanthropic initiatives (PHILi,t) and in-kind contributions to total corporate giving (INCi,t). The reported results are similar with Table 3. The findings reported in Table 3 are presented below according to their importance. Firms with high leverage (GEARi,t-1) tend to spend less on charity. Higher financial obligations are likely to be accompanied by stricter debt covenants, which would serve as a strong monitoring device and would reduce funds available for philanthropy. For each dollar increase in lagged leverage (GEARi,t-1), corporate giving decreases by 2.583. Larger firms (lnAi,t-1) carry a positive coefficient, suggesting that larger market visibility is related to higher charitable spending as a means to positively influence market participants‟ perceptions.

18

ACCEPTED MANUSCRIPT A higher percentage of independent directors sitting on the board of directors (IDBi,t-1) and a higher ratio of non-executive to executive directors (NEXECi,t-1) are negatively related to charitable giving. This suggests that the presence of effective

T

corporate governance mechanisms would monitor the managerial use of slack

RI P

resources and would seek to direct them to more efficient uses. Similar considerations hold for firms that are audited by a big 4 auditor (BIGAUi,t-1). Higher profitability (ROAi,t-1) reflects a higher potential surplus, which could be used to support firm ROAi,t-1, corporate giving increases by 0.749.

SC

strategic activities including charitable spending. For each dollar increase in lagged

MA NU

In line with Weisbrod and Dominguez (1986) and Frumkin and Kim (2001), fundraising (lnFREi,t-1) is similar to advertizing and exhibits a positive relationship with corporate giving. The ratio of in-kind contributions to total corporate giving (INCi,t) displays a significantly positive association with corporate giving. Likewise, firms that adopt structured philanthropic initiatives (PHILi,t) are more likely to display

ED

higher corporate giving. The use of structured philanthropic initiatives and of in-kind charitable activities would in general reduce agency costs (see Jensen, 1986). Overall,

PT

corporate giving is 0.205 greater for structured philanthropy (PHILi,t) as opposed to situations where structured philanthropic initiatives are absent. Also, for each dollar

CE

increase in in-kind contributions (INCi,t), corporate giving increases by 0.417. A high percentage of shares closely held by strategic investors, such as families, trusts, foundations and institutional investors (CLSi,t-1) exhibits a negative

AC

association with corporate giving, suggesting that their significant holdings would exercise more intensive monitoring on managers‟ charitable spending choices (see Brown et al, 2006). The same, but to a weaker extent, holds when a high proportion of ordinary shares is held by shareholders, who own 5% or more of the company (SOi,t1).

It is noteworthy that a new CEO or CFO (MCi,t-1) is likely to undertake charitable

activities possibly in order to establish a strong social profile. Net assets scaled by total revenue (NAi,t-1) displays a positive association with corporate giving. High net assets would give firms the ability to sufficiently support their business and charitable activities even in the case of a temporary decline in revenues. One dollar increase in net assets (NAi,t-1) leads to higher corporate giving by 0.039. The positive coefficient that is obtained for Tobin’s Q (TQi,t-1) reflects a negative association between corporate giving and agency costs. This is supported by

19

ACCEPTED MANUSCRIPT the adoption of structured philanthropic initiatives and the use of in-kind contributions, which would tend to affect agency costs downwards. Firms with significant advertizing expenses (ADVERTi,t-1) and R&D

T

expenditure (R&Di,t-1) demonstrate higher levels of corporate giving. For each dollar 1),

RI P

increase in lagged advertizing expenses (ADVERTi,t-1) and R&D expenditure (R&Di,tcorporate giving increases only by 0.002 and 0.008 respectively. Further, corporate

giving is positively related to bonuses that stem from prior year high performance

SC

(REMi,t-1). This would also indicate that the provision of bonuses would increase managerial productivity and firm profitability and would reinforce the potential for

MA NU

further corporate giving in the future. The findings show that utilities, petroleum, pharmaceutical and manufacturing firms tend to spend more on charitable activities. In contrast, financial firms and firms in mining, construction and retail donate less. Firms that exhibit a significant growth potential (MBVi,t-1) would tend to spend less on charitable activities in order to support and finance their growth options.

ED

According to Kasznik and Lev (1995) and Rogers and Stocken (2005), the probability of a firm being sued becomes higher in the case of a negative market response to its

PT

earnings announcement. The findings presented here indicate that firms more likely to face litigation (LITi,t-1) tend to give more on charity in an effort to favorably influence

CE

jurors‟ perceptions and courts‟ decisions. Finally, firms with higher liquidity (CFLOWi,t-1) tend to give more. However, for each dollar increase in lagged liquidity

AC

(CFLOWi,t-1), corporate giving increases only by 0.001. 5.3 Corporate Giving, Investor Perceptions and Value Relevance: Is Corporate Giving Worthwhile? The findings presented in Table 4 give support to H2, suggesting that corporate giving is value relevant and positively influences investor perceptions and future profitability and growth. Appendix 2 presents the findings without structured philanthropic initiatives (PHILi,t) and in-kind contributions to total corporate giving (INCi,t). The reported results are similar with Table 4. The findings reported in Table 4 are presented below according to their importance. Panel A also shows that firms that engage in charitable activities and adopt structured philanthropic initiatives tend to exhibit a positive association between lagged liquidity (CFLOWi,t-1), which enabled firms to spend on charity in the current year, and current stock price (Pi,t), as shown by the positive coefficients of CFLOWi,t-1 20

ACCEPTED MANUSCRIPT x CGi,t and CFLOWi,t-1 x PHILi,t respectively. Firms that spend on charity (CGi,t) and that adopt structured philanthropic initiatives (PHILi,t) would be positively valued by investors. For each dollar increase in corporate giving (CGi,t), the stock price

T

increases by 0.048. Also, the stock price is 0.128 greater for structured philanthropy

RI P

(PHILi,t) as opposed to situations where structured philanthropic initiatives are absent. Finally, lagged abnormal earnings (AEPSi,t-1) displays a positive coefficient, indicating that abnormal earnings would positively affect stock valuation. For each

SC

dollar increase in lagged abnormal earnings (AEPSi,t-1), the stock price increases by 0.018.

MA NU

The positive coefficients of EAR/BVi,t-1 x CGi,t, EAR/BVi,t-1 x PHILi,t, CFLOWi,t-1 x CGi,t and CFLOWi,t-1 x PHILi,t presented in Panel B show that firms that spend on charity and adopt structured philanthropic initiatives tend to present a positive association between their lagged earnings and lagged cash flows, which supported their charitable activities in the current year, with current firm value (see

ED

also Cormier and Magnan, 2007). Corporate giving (CGi,t) and the adoption of structured philanthropic initiatives (PHILi,t) carry positive coefficients, reflecting a

PT

positive relationship with firm stock market value. For each dollar increase in corporate giving (CGi,t), the market to book value of equity increases by 0.049. Also,

CE

the market to book value of equity is 0.148 greater for structured philanthropy (PHILi,t) as opposed to situations where structured philanthropic initiatives are absent. Panel C shows that corporate giving is negatively related to analyst forecast

AC

error (FE) and positively – although weakly – to analyst coverage (logAF). The provision of richer information about corporate giving would reduce the cost of following a firm and would further increase analyst following (see Houston et al, 2010). Especially in a monitored setting, such as the US, this would reduce information asymmetry and managerial opportunism and would subsequently increase forecast accuracy (see Basu, 1997) and reduce forecast dispersion (FD) (Horton et al, 2013). Corporate giving exhibits a positive relation with foreign sales (INTERi,t+1) in the subsequent period. Corporate giving is associated with lower return variability (RV) and bid-ask spreads (BA), reflecting lower risk levels. This is in consistency with the findings presented above, as highly liquid stocks would have low bid-ask spreads. In contrast, thinly traded stocks would display high bid-ask spreads. Panel C indicates that the positive news return variable in the subsequent period is significant (Ri,t+1 x 21

ACCEPTED MANUSCRIPT Posi,t+1), implying that although corporate giving has led to cash outflows in the year it has been incurred, the improvement of the firm‟s social image has favorable implications for the firm in the long-run. The findings show a positive coefficient for

T

trading volume (ST) and change in trading volume (CTV), reflecting a positive

RI P

relationship with corporate giving.

Corporate giving exhibits a positive relation with return on assets (ROAi,t+1) and market to book value of equity (MBVi,t+1) as an indicator of future growth in the

SC

subsequent period. Likewise, the findings provide evidence of a positive association between corporate giving and the Janis-Fadner coefficient (JF), suggesting that firms

MA NU

that spend on charity influence investor perceptions in a positive manner. Finally, revision optimism (RO) carries a positive coefficient, suggesting that corporate giving is hailed and is accompanied by upward revisions.

5.4 Corporate Giving and Discretionary Accruals

ED

The findings presented in Table 5 support H3 and show that firms that engage in charitable giving are likely to exhibit lower discretionary accruals. Appendix 3

PT

presents the findings without structured philanthropic initiatives (PHILi,t) and in-kind contributions to total corporate giving (INCi,t). The reported results are similar with

CE

Table 5. To further validate the results documented in Table 5, Appendix 4 presents a discretionary accruals comparison between givers and non-givers. The findings reported in Table 5 are presented below according to their importance.

AC

A higher market to book value of equity, which would reflect higher growth

potential and financial prospects, might urge firms to direct slack resources to discretionary and self-defined charitable causes. Here, the market to book value interaction variable (CGi,t x MBVi,t-1) carries a positive coefficient reflecting the positive association between growth and discretionary accruals (see also Cowan et al, 2012). Corporate giving (CGi,t) displays a negative association with earnings manipulation. The adoption of structured philanthropic initiatives (PHILi,t) and inkind contributions (INCi,t) are also negatively, but to a weaker extent, related to abnormal accruals. The donation of cash may motivate firms to mitigate the adverse impact of (charitable) spending on their liquidity using earnings manipulation techniques. Thus, it may be suggested that in-kind contributions may serve as defense against opportunism when a firm decides to undertake charitable activities. For each 22

ACCEPTED MANUSCRIPT dollar increase in corporate giving (CGi,t), abnormal accruals decrease by 0.580. Likewise, abnormal accruals is 0.116 less for structured philanthropy (PHILi,t) as opposed to situations where structured philanthropic initiatives are absent. Also, for

T

each dollar increase in in-kind contributions (INCi,t), abnormal accruals decrease by

RI P

0.056. Fundraising expenses (lnFREi,t-1) demonstrate a negative association with earnings manipulation. Fundraising behaves similarly to advertizing since it provides useful information about the cause and the funding structure. The communication of

SC

such information would positively affect charitable giving and would lead to lower opportunism (Frumkin and Kim, 2001).

MA NU

The presence of a higher percentage of shares closely held by strategic investors (CLSi,t-1), such as families, trusts, foundations and institutional investors, and being audited by a big auditor (BIGAUi,t-1) would monitor managers‟ decisions more intensively and would thus reduce the scope for earnings manipulation. The percentage of independent directors sitting on the board of directors (IDBi,t-1) and the

ED

ratio of non-executive to executive directors (NEXECi,t-1) also display a negative relationship with earnings manipulation.

PT

As hypothesized, firms that spend on charity tend to exhibit less earnings manipulation as shown by the negative coefficient of BELi,t-1 x CGi,t and the positive

CE

coefficient of HIGHi,t-1 x CGi,t, which suggest that charity givers with low (high) premanaged earnings do not manipulate earnings upwards (downwards). Earnings manipulation is negatively associated with firm size (CGi,t x lnAi,t-1), suggesting that

AC

corporate giving of large firms would attract market participants‟ attention and need for detailed information and would thus limit the expression of opportunistic behaviors. The positive coefficient of CGi,t x CFLOWi,t-1 indicates that charity givers with low liquidity would display low abnormal accruals. The results presented in Appendix 4 are similar with Table 5. Appendix 4 shows that corporate givers (dCGi,t = 1) display a negative association with abnormal accruals (AAi,t-1). As also shown in Table 5, they exhibit a negative coefficient for BELi,t-1 x dCGi,t and a positive coefficient for HIGHi,t-1 x dCGi,t. Further, corporate givers display a negative coefficient for dCGi,t x lnAi,t-1 and a positive coefficient for dCGi,t x CFLOWi,t-1. Corporate governance variables, i.e. IDBi,t-1, NEXECi,t-1 and BIGAUi,t-1, are also found to carry negative coefficients.

23

ACCEPTED MANUSCRIPT 6 Conclusions This study focuses on corporate philanthropy and examines the financial attributes of firms that engage in charitable activities. It is hypothesized that certain financial

T

characteristics, such as in-kind contributions, size, liquidity and the possibility of

RI P

litigation, would encourage corporate giving, while others, such as leverage, agency costs, growth and corporate governance, would discourage such practices. This study also assesses the value relevance of corporate giving, and investigates the impact of

SC

giving on investor perceptions and future profitability and growth. Finally, this paper examines the association between charitable spending and earnings manipulation.

MA NU

High leverage and strong corporate governance mechanisms tend to monitor the use of slack resources more intensively and to lead to lower (and/or higher quality) charitable spending. Larger and profitable firms tend to give more, while those that are audited by a big auditor tend to engage in lower charitable spending. The adoption of structured philanthropic initiatives and the use of in-kind

ED

contributions would encourage corporate giving as it would be expected to reduce agency costs. This study has found that a new CEO or CFO is more likely to spend on

PT

charity in order to influence market participants‟ perceptions. Corporate giving is value relevant and positively influences investor

CE

perceptions. This holds especially when firms adopt structured philanthropic initiatives. Corporate giving is found to be negatively related to analyst forecast error and dispersion and positively to analyst coverage. Also, it is associated with lower

AC

bid-ask spreads. The long-term dimension of corporate giving appears to overshadow the outflows incurred and the resources committed in the short run, and leads to positive returns in the current and subsequent periods. Firms that engage in charitable giving tend to exhibit lower earnings manipulation. However, higher growth prospects may urge firms to direct slack resources to discretionary charitable causes and thus may give rise to earnings manipulation. In-kind contributions are negatively related to managerial opportunism. Likewise, strong corporate governance and being audited by a big auditor tend to reduce the potential for earnings manipulation. This is a pioneer study as it is the first to link corporate giving to earnings manipulation and value relevance. It also contributes to the analysts‟ forecast literature in relation to corporate giving. The findings are useful for investors, financial analysts and other market participants as they shed light on the financial 24

ACCEPTED MANUSCRIPT attributes of charitable giving covering a wide spectrum of different income statement and balance sheet items. A major financial event, such as a change in tax rules, a lawsuit, or an economic crisis, might encourage earnings management phenomena

T

(see Huang et al, 2009), which might in turn reduce firm value. Here, the study

RI P

contributes by providing evidence that charitable spending has reduced the scope for earnings manipulation and has improved investor perceptions, stock valuation and future company financial performance. Another contribution of this study is that

SC

although firms with low liquidity and high leverage would be tempted to engage in earnings manipulation practices in order to influence their liquidity upwards and meet

MA NU

their debt covenants and financial commitments, the public scrutiny and visibility of charitable giving leads to charity givers displaying lower abnormal accruals. This study implies that effective corporate governance would restrain unwelcome manipulative managerial behaviors and would therefore reduce agency costs and strengthen firm value. To this end, accounting regulation should reinforce

ED

the quality of disclosure requirements that relate to fundraising and corporate giving in order to identify opportunistic behaviors that may be hidden under „noble‟

PT

charitable causes and smooth the relationship between managers and stakeholders. This should be stressed especially for firms with high growth potential, since they

CE

may be tempted to direct financial surpluses to self-defined causes in order to influence capital providers and their terms of financing. Hence, stricter corporate giving-specific disclosure requirements would further improve forecast accuracy,

AC

analyst coverage and analyst revisions, which, as reported above, already are positively affected by charitable giving. Further, this study has significant implications for the litigation literature, since charitable giving may be intended to serve opportunistic objectives and influence jurors and judges. This study also implies that in-kind contributions, as opposed to cash contributions, may be used as a governance or monitoring tool to reduce the scope for earnings manipulation. Considering that corporate giving should be closely linked to information value, future research should examine the extent to which corporate giving increases earnings quality and promotes earnings conservatism. This should subsequently call for an investigation of the association between corporate giving and the cost of capital. Future research should study the tax-related considerations and motivation for charitable giving. It should also investigate the potential for earnings manipulation that may be applied under different tax settings in order to maintain firms‟ ability to 25

ACCEPTED MANUSCRIPT spend on charity and meet personal managerial objectives. Future research should examine how corporate giving has been affected by the European debt crisis. It should also assess the corporate giving-related decision-making process of shareholders or

T

strategic investors that hold a significant proportion of ordinary shares. Future

RI P

research should comparatively examine corporate giving in countries with different institutional characteristics, such as common-law and code-law countries, in emerging markets, and in countries where the quality of financial reporting is perceived to be

SC

lower.

MA NU

References

Aerts, W. and D. Cormier (2009), “Media Legitimacy and Corporate Environmental Communication”, Accounting, Organizations and Society, 34 (1), pp. 1-27. Aggarwal, R., M. Evans and D. Nanda (2012), “Nonprofit Boards: Size, Performance

AC

CE

PT

ED

and Managerial Incentives”, Journal of Accounting and Economics, 53 (1-2), pp. 466-487. Amato, L. and C. Amato (2007), “The Effects of Firm Size and Industry on Corporate Giving”, Journal of Business Ethics, 72 (3), pp. 229-241. Amihud, Y. (2002), “Illiquidity and Stock Returns”, Journal of Financial Markets, 5, pp. 31-56. Amit, R. and P. Schoemaker (1993), “Strategic Assets and Organizational Rent”, Strategic Management Journal, 14 (1), pp. 33-46. Barth, M. and A. Hutton (2004), “Analyst Earnings Forecast Revisions and the Pricing of Accruals”, Review of Accounting Studies, 9, pp. 59-96. Bartkus, B., S. Morris and B. Seifert (2002), “Governance and Corporate Philanthropy”, Business and Society, 41, pp. 319-344. Basu, S. (1997), “The Conservatism Principle and the Asymmetric Timeliness of Earnings”, Journal of Accounting and Economics, 24, pp. 3-37. Berman, S., A. Wicks, S. Kotha and T. Jones (1999), “Does Stakeholder Orientation Matter? The Relationship between Stakeholder Management Models and Firm Financial Performance”, Academy of Management Journal, 42 (5), pp. 488-506. Bhagat, S. and R. Jefferis (2002), “The Econometrics of Corporate Governance Studies”, MIT Press, Cambridge. Boatsman, J. and S. Gupta (1996), “Taxes and Corporate Charity: Empirical Evidence from Micro-Level Panel Data”, National Tax Journal, 49, pp. 193-213. Brammer, S. and A. Millington (2005), “Corporate Reputation and Philanthropy: An Empirical Analysis”, Journal of Business Ethics, 61 (1), pp. 29-44. Brown, W., E. Helland and J. Smith (2006), “Corporate Philanthropic Practices”, Journal of Corporate Finance, 12 (5), pp. 855-877. Buchholtz, A., A. Amason and M. Rutherford (1999), “Beyond Resources: The Mediating Effects of Top Management Discretion and Values on Corporate Philanthropy”, Business & Society, 38 (2), pp. 167-187. Carroll, A. (1991), “The Pyramid of Corporate Social Responsibility: Toward the Moral Management of Organizational Stakeholders”, Business Horizons, 34 (4), pp. 39-48.

26

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

Clarkson, M. (1995), “A Stakeholder Framework for Analyzing and Evaluating Corporate Social Performance”, Academy of Management Review, 20 (1), pp. 92-117. Clarkson, P., X. Fang, Y. Li and G. Richardson (2013), “The Relevance of Environmental Disclosures: Are Such Disclosures Incrementally Informative?”, Journal of Accounting and Public Policy, 32 (5), pp. 410-431.. Cormier, D. and M. Magnan (2007), “The Revisited Contribution of Environmental Reporting to Investors‟ Valuation of a Firm‟s Earnings: An International Perspective”, Ecological Economics, 62 (3-4), pp. 613-626. Cotter, J., A. Tarca and M. Wee (2012), “IFRS Adoption and Analysts‟ Earnings Forecasts: Australian Evidence”, Accounting and Finance, 52 (2), pp. 395-419. Covrig, V., M. Defond and M. Hung (2007), “Home Bias, Foreign Mutual Fund Holdings, and the Voluntary Adoption of International Accounting Standards”, Journal of Accounting Research, 45, pp. 41-70. Cowan, A., C. Huang, P. Padmanabhan and C. Wang (2012), “The Determinants of Foreign Giving: An Exploratory Empirical Investigation of US Manufacturing Firms”, International Business Review, 22 (2), pp. 407-420. Daske, H., L. Hail, C. Leuz and R. Verdi (2013), “Adopting a Label: Heterogeneity in the Economic Consequences around IAS/ IFRS Adoptions”, Journal of Accounting Research, 51 (3), pp. 495-547. Dechow, P. and R. Sloan (1995), “Detecting Earnings Management”, The Accounting Review, 70, pp. 193-225. Dhaliwal, D., O. Li, A. Tsang and G. Yang (2011), “Voluntary Non-Financial Disclosure and the Cost of Equity Capital: The Initiation of Corporate Social Responsibility Reporting”, The Accounting Review, 86 (1), pp. 59-100. Ding, Y., O. Hope, T. Jeanjean and H. Stolowy (2007), “Differences between Domestic Accounting Standards and IAS: Measurement, Determinants and Implications”, Journal of Accounting and Public Policy, 26 (1), pp. 1-38. Easton, P. (2004), “PE Ratios, PEG Ratios, and Estimating the Implied Expected Rate of Returns on Equity Capital”, Accounting Review, 79, pp. 73-95. Elgers, P., M. Lo and R. Pfeiffer (2001), “Delayed Security Price Adjustments to Financial Analysts‟ Forecasts of Annual Earnings”, The Accounting Review, 76, pp. 613-632. Elliott, J. and D. Philbrick (1990), “Accounting Changes and Earnings Predictability”, The Accounting Review, 65 (1), pp. 157-174. Fombrun, C. and M. Shanley (1990), “What‟s In a Name? Reputation Building and Corporate Strategy”, Academy of Management Journal, 33, pp. 233-258. Frumkin, P. and M. Kim (2001), “Strategic Positioning and the Financing of Nonprofit Organizations: Is Efficiency Rewarded in the Contributions Marketplace?”, Public Administration Review, 61 (3), pp. 266-275. Galaskiewicz, J. (1985), “Social Organization of an Urban Grants Economy: A Study of Business Philanthropy and Nonprofit Organizations”, Academic Press, Orlando, FL. Galaskiewicz, J. (1997), “An Urban Grants Economy Revisited: Corporate Charitable Contributions in the Twin Cities, 1979-81, 1987-89”, Administrative Science Quarterly, 42 (3), pp. 445-471. Godfrey, P. (2005), “The Relationship between Corporate Philanthropy and Shareholder Wealth: A Risk Management Perspective”, Academy of Management Review, 30 (4), pp. 777-798.

27

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

Gordon, T. and S. Khumawala (1999), “The Demand for Nonprofit Financial Statements: A Model for Individual Giving”, Journal of Accounting Literature, 18, pp. 31-56. Greenlee, J. and J. Trussel (2000), “Predicting the Financial Vulnerability of Charitable Organizations”, Nonprofit Management and Leadership, 11 (2), pp. 199-210. Haley, U. (1991), “Corporate Contributions as Managerial Masques: Reframing Corporate Contributions as Strategies to Influence Society”, Journal of Management Studies, 28 (5), pp. 485-509. Harvey, C., K. Lins and A. Roper (2004), “The Effect of Capital Structure when Expected Agency Costs are Extreme”, Journal of Financial Economics, 74, pp. 3-30. Healy, P., A. Hutton and K. Palepu (1999), “Stock Price Performance and Intermediation Changes Surrounding Sustained Increases in Disclosure”, Contemporary Accounting Research (Fall), pp. 485-520. Healy, P. and K. Palepu (2001), “Information Asymmetry, Corporate Disclosure, and the Capital Markets: A Review of the Empirical Disclosure Literature”, Journal of Accounting and Economics, 31, pp. 405-440. Horton, J., G. Serafeim and I. Serafeim (2008), “Does Mandatory IFRS Adoption Improve the Information Environment?”, Working Paper, Available at: http://www.lse.ac.uk/accounting/news/MAFG/Serafeimpaper.pdf. Horton, J., G. Serafeim and I. Serafeim (2013), “Does Mandatory IFRS Adoption Improve the Information Environment?”, Contemporary Accounting Research, 30 (1), pp. 388-423. Houston, J., B. Lev and J. Tucker (2010), “To Guide or Not to Guide? Causes and Consequences of Stopping and Subsequently Resuming Earnings Guidance”, Contemporary Accounting Research, 27 (1), pp. 143-185. Huang, P., Y. Zhang, D. Deis and J. Moffitt (2009), “Do Artificial Income Smoothing and Real Income Smoothing Contribute to Firm Value Equivalently?”, Journal of Banking & Finance, 33 (2), pp. 224-233. Janis, I. and R. Fadner (1965), “The Coefficient of Imbalance”, in Lasswell, H. and N. Leites and Associates (Eds.), “Language of Politics”, Cambridge, MA: MIT Press, pp. 153-169. Jensen, M. and W. Meckling (1976), “Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure”, Journal of Financial Economics, 3, pp. 305-360. Jensen, M. (1986), “Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers”, American Economic Review, 76, pp. 323-329. Johanson, J. and J. Vahlne (2006), “Commitment and Opportunity Development in the Internationalization Process: A Note on the Uppsala Internationalization Process Model”, Management International Review, 46 (2), pp. 165-178. Kasznik, R. and B. Lev (1995), “To Warn or Not to Warn: Management Disclosure in the Face of an Earnings Surprise”, The Accounting Review, 70 (1), pp. 113-134. Leuz, C. and R. Verrecchia (2000), “The Economic Consequences of Increased Disclosure”, Journal of Accounting Research, 38, pp. 91–124. Leuz, C., D. Nanda and P. Wysocki (2003), “Earnings Management and Investor Protection: An International Comparison”, Journal of Financial Economics, 69 (3), pp. 505-528. Louis, H. and D. Robinson (2005), “Do Managers Credibly Use Accruals to Signal Private Information? Evidence from the Pricing of Discretionary Accruals around Stock Splits”, Journal of Accounting and Economics, 39, pp. 361-380. 28

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

Margolis, J. and J. Walsh. 2003. “Misery Loves Companies: Rethinking Social Initiatives by Business”, Administrative Sciences Quarterly, 48: pp. 268-305. Marquis, C., M. Glynn and G. Davis (2007), “Community Isomorphism and Corporate Social Action”, Academy of Management Review, 32, pp. 925-945. McGuire, J., A. Sundgren and T. Schneeweis (1988), “Corporate Social Responsibility and Firm Financial Performance”, Academy of Management Journal, 31 (4), pp. 854-872. Mescon, T. and D. Tilson (1987), “Corporate Philanthropy: A Strategic Approach to the Bottom-Line”, California Management Review, 29 (2), pp. 49-61. Mitchell, R., B. Agle and D. Wood (1997), “Toward a Theory of Stakeholder Identification and Salience: Defining the Principle of Who and What Really Counts”, Academy of Management Review, 22 (4), pp. 853-886. Muller, K., E. Riedl and T. Sellhorn (2008), “Consequences of Voluntary and Mandatory Fair Value Accounting: Evidence Surrounding IFRS Adoption in the EU Real Estate Industry”, Working Paper, Available at: http://www.hbs.edu/faculty/Publication%20Files/09-033_14a96a28-1bd8-4fbd9313-b0ad89e1ea63.pdf. O‟Neill, H., C. Saunders and A. McCarthy (1989), “Board Members, Corporate Social Responsiveness and Profitability: Are Tradeoffs Necessary?”, Journal of Business Ethics, 8 (5), pp. 353-357. Ohlson, J. (1995), “Earnings, Book Values, and Dividends in Equity Valuation”, Contemporary Accounting Research, 11 (2), pp. 661-687. Orlitzky, M., F. Schmidt and S. Rynes (2003), “Corporate Social and Financial Performance: A Meta-Analysis”, Organization Studies, 24 (3), pp. 403-441. Pallot, J. (1990), “The Nature of Public Assets: A Response to Mautz”, Accounting Horizons, 4 (2), pp. 79-85. Parsons, L. and J. Trussel (2008), “Fundamental Analysis of Not-For-Profit Financial Statements: An Examination of Financial Vulnerability Measures”, Research in Governmental and Nonprofit Accounting, 12, pp. 35-56. Peasnell, K., P. Pope and S. Young (2005), “Board Monitoring and Earnings Management: Do Outside Directors Influence Abnormal Accruals?”, Journal of Business Finance and Accounting, 32 (7-8), pp. 1311-1346. Piotroski, J. and D. Roulstone (2004), “The Influence of Analysts, Institutional Investors, and Insiders on the Incorporation of Market, Industry, and FirmSpecific Information into Stock Prices”, The Accounting Review, 79, pp. 11191151. Plumlee, M., D. Brown and S. Marshall (2010), “Voluntary Environmental Disclosure Quality and Firm Value: Further Evidence”, Working Paper, Available at: http://www2.business.umt.edu/seminar/draft_Montana.pdf. Roberts, P. and G. Dowling (2002), “Corporate Reputation and Sustained Superior Financial Performance”, Strategic Management Journal, 23, pp. 1077-1093. Rogers, J. and P. Stocken (2005), “Credibility of Management Forecasts”, The Accounting Review, 80 (4), pp. 1233-1260. Rowley, T. and S. Berman (2000), “A Brand New Brand of Corporate Social Performance”, Business & Society, 39 (4), pp. 397-418. Sanchez, C. (2000), “Motives for Corporate Philanthropy in El Salvador: Altruism and Political Legitimacy”, Journal of Business Ethics, 27 (4), pp. 363-375. Seifert, B., S. Morris and B. Bartkus (2003), “Comparing Big Givers and Small Givers: Financial Correlates of Corporate Philanthropy”, Journal of Business Ethics, 45 (3), pp. 195-211.

29

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

Tinkelman, D. (1999), “Factors Affecting the Relation between Donations to Nonprofit Organizations and an Efficiency Ratio”, Research in Government and Nonprofit Accounting, 10, pp. 135-161. Trussel, J. and J. Greenlee (2004), “A Financial Rating System for Charitable Nonprofit Organizations”, Research in Government and Nonprofit Accounting, 11, pp. 93-116. Trussel, J. and L. Parsons (2008), “Financial Reporting Factors Affecting Donations to Charitable Organizations”, Advances in Accounting, 23, pp. 263-285. Useem, M. (1988), “Market and Institutional Factors in Corporate Contributions”, California Management Review, 30 (2), pp. 77-88. Vaidyanathan, B. (2008), “Science of Generosity; Corporate Giving: A Literature Review”, Center for the Study of Religion and Society, University of Notre Dame. Vogel, D. (2005), “The Market for Virtue: The Potential and Limits of Corporate Social Responsibility”, Washington DC: The Brookings Institution. Waddock, S. and S. Graves (1997), “The Corporate Social Performance-Financial Performance Link”, Strategic Management Journal, 18 (4), pp. 303-319. Wang, H., J. Choi and J. Li (2008), “Too Little or Too Much? Untangling the Relationship between Corporate Philanthropy and Firm Financial Performance”, Organization Science, 19 (1), pp. 143-159. Weisbrod, B. and N. Dominguez (1986), “Demand for Collective Goods in Private Nonprofit Markets: Can Fundraising Expenditures Help Overcome Free-Rider Behavior?”, Journal of Public Economics, 30, pp. 83-96. Wood, D. and R. Jones (1995), “Stakeholder Mismatching: A Theoretical Problem in Empirical Research on Corporate Social Performance”, International Journal of Organizational Analysis, 3 (3), pp. 229-267. Zhang, R., J. Zhu, H. Yue and C. Zhu (2010), “Corporate Philanthropic Giving, Advertising Intensity, and Industry Competition Level”, Journal of Business Ethics, 94 (1), pp. 39-52.

30

ACCEPTED MANUSCRIPT Table 1 Corporate Giving in 2012: Top 10 Givers Company

Contribution

Share of 2011 pre-tax profits donated in 2012 1.3%

Wells Fargo & Company

$315,845,766 (cash)

Walmart Stores

$311,607,280 (cash) $755,868,381 (products)

Chevron Corporation

$262,430,000 (cash)

Goldman Sachs Group

$241,278,912 (cash)

Exxon Mobil Corporation

$213,374,183 (cash) $2,433,200 (products)

Bank of America

$222,862,368 (cash)

JPMorgan Chase & Company

$183,471,434 (cash)

General Electric

$161,500,000 (cash)

Target Corporation

$147,038,722 (cash) $76,554,400 (products)

5.0%

Citigroup

$137,032,650 (cash)

0.9%

MA NU

SC

RI P

T

4.5%

Source: The Chronicle of Philanthropy (July 14, 2013) http://philanthropy.com/article/10-Companies-That-Gave-the/140261/ * Due to money loss in 2011, this variable could not be determined.

ED

Table 2 Descriptive Statistics Panel A Descriptive Statistics for the Entire Sample Variables Mean CGi,t (per million $ assets) 822 lnFREi,t-1 7.503 ROAi,t-1 0.071 GEARi,t-1 0.531 lnAi,t-1 14.123 CFLOWi,t-1 0.067 MBVi,t-1 3.283 ADVERTi,t-1 0.032 R&Di,t-1 0.046 IDBi,t-1 0.442 CLSi,t-1 0.238 FE 0.022 ST 0.518 BA 0.019 AAi,t-1 0.093 Panel B Descriptive Statistics per Sample Year CGi,t (per million $ assets) 2008 Mean 809 25th quartile 156 Median 336 75th quartile 975 2009 Mean 815 25th quartile 164 Median 344 75th quartile 1003 2010 Mean 824 25th quartile 172 Median 358 75th quartile 986 2011 Mean 829 25th quartile 159 Median 364 75th quartile 994

AC

CE

PT

Std Deviation 355 4.312 0.739 0.200 2.901 0.231 4.097 0.053 0.061 0.131 0.148 0.051 0.765 0.027 0.225

31

0.6% 3.9% 0.3% N/A* 0.7% 0.8%

ACCEPTED MANUSCRIPT

SC

RI P

T

2012 Mean 833 25th quartile 167 Median 372 75th quartile 1086 2008-2012 822 The sample period is 2008 to 2012. The sample consists of 687 firms. CGi,t is annual corporate giving scaled by total assets. lnFREi,t-1 is the natural logarithm of fundraising expenses. ROAi,t-1 is income before extraordinary items scaled by total assets. GEARi,t-1 is debt scaled by total assets. lnAi,t-1 is the natural logarithm of total assets. CFLOWi,t-1 is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. MBVi,t-1 is market to book value of equity. ADVERTi,t-1 is advertizing expenses scaled by total sales. R&Di,t-1 is research and development expenditure scaled by total sales. IDBi,t-1 is the percentage of independent directors sitting on the board of directors. CLSi,t-1 is the percentage of shares closely held by strategic investors. FE is analyst forecast error, i.e. mean minus actual forecast, scaled by absolute actual earnings. ST is trading volume scaled by market value of outstanding equity. BA is the bid-ask spread. AAi,t-1 is abnormal accruals and obtained as shown in equation (10).

INCi,t lnFREi,t-1 ROAi,t-1

TQi,t-1 CFLOWi,t-1 MBVi,t-1

INDi,t-1 IDBi,t-1 NEXECi,t-1 MCi,t-1 REMi,t-1 BIGAUi,t-1 NAi,t-1 SOi,t-1 CLSi,t-1 LITi,t-1 Constant R2 adj.

AC

ADVERTi,t-1

CE

lnAi,t-1

PT

ED

GEARi,t-1

R&Di,t-1

Table 3 Financial Attributes of Corporate Giving Coefficients 0.205*** (0.023) 0.417*** (0.079) 0.612*** (0.025) 0.749*** (0.063) -2.583*** -2.583 (0.898) 1.130*** (0.067) 0.016*** (0.006) 0.001** (0.0008) -1.059* (0.611) 0.002*** (0.001) 0.008*** (0.002) 0.017** (0.007) -0.996*** (0.053) -0.581*** (0.161) 0.067*** (0.019) 0.032** (0.013) -0.232*** (0.046) 0.0394*** (0.016) -0.160* (0.094) -0.077*** (0.010) 0.192* (0.115) 1.854 (1.885) 0.647

MA NU

Variables PHILi,t

32

ACCEPTED MANUSCRIPT Sample size

N=3,435

SC

RI P

T

***, ** and * indicate statistical significance at the 1%, 5% and 10% level (two-tailed) respectively. The standard error is in parentheses. The dependent variable is CGi,t, which is annual corporate giving scaled by total assets. PHILi,t = 1 for firms that adopt structured philanthropic initiatives and PHILi,t = 0 otherwise. INCi,t is a firm‟s in-kind contributions as a proportion to total corporate giving. lnFREi,t-1 is the natural logarithm of fundraising expenses. ROAi,t-1 is income before extraordinary items scaled by total assets. GEARi,t-1 is debt scaled by total assets. lnAi,t-1 is the natural logarithm of total assets. TQi,t-1 is equal to market value of common equity plus book value of preferred stock, plus book value of long-term debt and current liabilities, scaled by book value of total assets. CFLOWi,t-1 is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. MBVi,t-1 is market to book value of equity. ADVERTi,t-1 is advertizing expenses scaled by total sales. R&Di,t-1 is research and development expenditure scaled by total sales. INDi,t-1 is a dummy variable and proxies for industry classification. IDBi,t-1 is the percentage of independent directors sitting on the board of directors. NEXECi,t-1 is the ratio of non-executive to executive directors. MCi,t-1 = 1 when changes in the management have occurred and MCi,t-1 = 0 otherwise. REMi,t-1 is performance-related bonus. BIGAUi,t-1 = 1 for firms that are audited by a big 4 auditor and BIGAUi,t-1 = 0 otherwise. NAi,t-1 is net assets scaled by total revenue. SOi,t-1 is the proportion of ordinary shares held by shareholders, who own 5% or more of the company. CLSi,t-1 is the percentage of shares closely held by strategic investors. LITi,t-1 is the probability of litigation and is estimated as shown in equation (2).

AC

CE

PT

ED

MA NU

Table 4 Corporate Giving, Investor Perceptions and Value Relevance Panel A Equation (3) Panel B Equation (5) Panel C Equation (6) Variables Coefficients Variables Coefficients Variables Coefficients BVi,t-1 0.011 1/BVi,t-1 0.018* FE -0.218*** (0.027) (0.010) (0.073) FD AEPSi,t-1 0.018** EAR/BVi,t-1 -0.001 -0.356*** (0.008) (0.002) (0.093) logAF CGi,t 0.048*** CGi,t 0.049** 0.035** (0.004) (0.022) (0.016) PHILi,t 0.128*** PHILi,t 0.148*** RO 0.002* (0.048) (0.009) (0.001) INCi,t 0.0018 INCi,t 0.006 ST 0.234*** (0.002) (0.008) (0.099) EAR/BVi,t-1 x CGi,t CFLOWi,t-1 0.0013*** 0.297*** CTV 0.331*** (0.0001) (0.033) (0.027) CFLOWi,t-1 x CGi,t 0.316*** EAR/BVi,t-1 x PHILi,t 0.177*** BA -0.428*** (0.020) (0.040) (0.155) CFLOWi,t-1 x PHILi,t 0.138*** CFLOWi,t-1 0.053*** Ri,t 0.086** (0.048) (0.010) (0.044) Constant 0.058 CFLOWi,t-1 x CGi,t 0.287*** Posi,t -0.082 (0.020) (0.078) (0.109) CFLOWi,t-1 x PHILi,t 0.187*** Ri,t x Posi,t 0.132 (0.018) (0.349) Constant 0.100 Ri,t+1 0.006*** (0.127) (0.001) Posi,t+1 0.171 (0.246) Ri,t+1 x Posi,t+1 0.361* (0.221) RV -0.553*** (0.115) JF 0.028*** (0.008) ROAi,t+1 0.107*** (0.010) MBVi,t+1 0.165*** (0.024) INTERi,t+1 0.729*** (0.015) Constant 0.163 (0.472) R2 adj. R2 adj. R2 adj. 0.767 0.617 0.520 Sample size Sample size Sample size N=3,435 N=3,435 N=3,435 ***, ** and * indicate statistical significance at the 1%, 5% and 10% level (two-tailed) respectively. The standard error is in parentheses. In Panel A, the dependent variable is Pi,t, which is the stock price at the end of the year. In Panel B, the dependent variable is MBVi,t, which is market to book value of equity. In Panel C, the dependent variable is CGi,t, which is annual corporate giving scaled by total assets. BVi,t-1 is book value of equity. AEPSi,t-1 is earnings per share at the end of the year, minus the cost of equity capital multiplied by book value of

33

ACCEPTED MANUSCRIPT

RI P

T

equity at the beginning of the year. CGi,t is annual corporate giving scaled by total assets. PHILi,t = 1 for firms that adopt structured philanthropic initiatives and PHILi,t = 0 otherwise. INCi,t is a firm‟s in-kind contributions as a proportion to total corporate giving. CFLOWi,t-1 is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. EAR/BVi,t-1 is net income before extraordinary items scaled by book value of equity. FE is analyst forecast error, i.e. mean minus actual forecast, scaled by absolute actual earnings. FD is the standard deviation of analyst forecasts scaled by absolute actual earnings. logAF is the log of one plus the number of analysts covering the firm. RO is revision optimism and is the difference between upward revisions and downward revisions scaled by the number of estimates. ST is trading volume scaled by market value of outstanding equity. CTV is the price change in trading volume. BA is the bid-ask spread. Ri,t, Ri,t+1 are stock returns. Pos = 1 for positive return values and Pos = 0 otherwise. RV is the standard deviation of returns. JF is the Janis-Fadner coefficient and obtained as shown in equations (7) and (8). ROAi,t+1 is income before extraordinary items scaled by total assets. MBVi,t+1 is market to book value of equity. INTERi,t+1 is foreign sales scaled by total sales.

SC

Table 5 Corporate Giving and Discretionary Accruals Coefficients -0.580*** (0.226) PHILi,t -0.116*** (0.046) INCi,t -0.056** (0.027) lnFREi,t-1 -0.502*** (0.173) BELi,t-1 -0.005 (0.007) HIGHi,t-1 0.134 (0.498) BELi,t-1 x CGi,t -0.197*** (0.034) HIGHi,t-1 x CGi,t 0.001*** (0.0003) ICFi,t-1 0.093 (0.060) lnAi,t-1 0.199 (0.141) CFLOWi,t-1 0.123*** (0.008) MBVi,t-1 0.653*** (0.098) CGi,t x lnAi,t-1 -0.038*** (0.014) CGi,t x CFLOWi,t-1 0.008* (0.004) CGi,t x MBVi,t-1 0.729*** (0.085) IDBi,t-1 -0.311* (0.165) NEXECi,t-1 -0.077** (0.034) BIGAUi,t-1 -0.203*** (0.027) CLSi,t-1 -0.443*** (0.048) Constant 0.157 (0.332) R2 adj. 0.702 Sample size N=3,435 ***, ** and * indicate statistical significance at the 1%, 5% and 10% level (two-tailed) respectively. The standard error is in parentheses. The dependent variable is AAi,t-1, which is abnormal accruals and obtained as shown in equation (10). CGi,t is annual corporate giving scaled by total assets. PHILi,t = 1 for firms that adopt structured philanthropic initiatives and PHILi,t = 0 otherwise. INCi,t is a firm‟s in-kind contributions as a proportion to total corporate giving. lnFREi,t-1 is the natural logarithm of fundraising expenses. BELi,t-1 = 1 if earnings minus abnormal accruals, scaled by total assets, is less than zero or less than lagged reported earnings and 0 otherwise.

AC

CE

PT

ED

MA NU

Variables CGi,t

34

ACCEPTED MANUSCRIPT

T

HIGHi,t-1 = 1 if earnings minus abnormal accruals, scaled by total assets, is more than zero or more than lagged reported earnings and 0 otherwise. ICFi,t-1 is the standard deviation of operating income scaled by total assets divided by the standard deviation of operating cash flow scaled by total assets. lnAi,t-1 is the natural logarithm of total assets. CFLOWi,t-1 is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. MBVi,t-1 is market to book value of equity. IDBi,t-1 is the percentage of independent directors sitting on the board of directors. NEXECi,t-1 is the ratio of non-executive to executive directors. BIGAUi,t-1 = 1 for firms that are audited by a big 4 auditor and BIGAUi,t-1 = 0 otherwise. CLSi,t-1 is the percentage of shares closely held by strategic investors.

AC

CE

PT

ED

MA NU

SC

RI P

Appendix 1 Financial Attributes of Corporate Giving without PHILi,t and INCi,t Variables Coefficients lnFREi,t-1 0.681*** (0.094) ROAi,t-1 0.111*** (0.014) GEARi,t-1 -2.761*** (0.555) lnAi,t-1 1.527*** (0.362) TQi,t-1 0.109** (0.046) CFLOWi,t-1 0.105*** (0.036) MBVi,t-1 -1.160*** (0.344) ADVERTi,t-1 0.066*** (0.010) R&Di,t-1 0.002*** (0.0003) INDi,t-1 0.001** (0.0008) IDBi,t-1 -0.810*** (0.179) NEXECi,t-1 -0.408* (0.243) MCi,t-1 0.256*** (0.018) REMi,t-1 0.04*** (0.004) BIGAUi,t-1 -0.501*** (0.060) NAi,t-1 0.001*** (0.0003) SOi,t-1 -0.552*** (0.004) CLSi,t-1 -0.035** (0.015) LITi,t-1 0.812* (0.482) Constant 0.775 (0.138) R2 adj. 0.589 Sample size N=3,435

***, ** and * indicate statistical significance at the 1%, 5% and 10% level (two-tailed) respectively. The standard error is in parentheses. The dependent variable is CGi,t, which is annual corporate giving scaled by total assets. lnFREi,t-1 is the natural logarithm of fundraising expenses. ROAi,t-1 is income before extraordinary items scaled by total assets. GEARi,t-1 is debt scaled by total assets. lnAi,t-1 is the natural logarithm of total assets. TQi,t-1 is equal to market value of common equity plus book value of preferred stock, plus book value of long-term debt and current liabilities, scaled by book value of total assets. CFLOWi,t-1 is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. MBVi,t-1 is market to book value of equity. ADVERTi,t-1 is advertizing expenses scaled by total sales. R&Di,t-1 is research and development expenditure scaled by total sales. INDi,t-1 is a dummy variable and proxies for industry classification. IDBi,t-1 is the percentage of independent directors sitting on the board of directors. NEXECi,t1 is the ratio of non-executive to executive directors. MCi,t-1 = 1 when changes in the management have occurred and MCi,t-1 = 0 otherwise. REMi,t-1 is performance-related bonus. BIGAUi,t-1 = 1 for firms that are audited by a big 4 auditor and BIGAUi,t-1 = 0 otherwise. NAi,t-1 is net assets scaled by total revenue. SOi,t-1 is the proportion of ordinary shares held by shareholders, who

35

ACCEPTED MANUSCRIPT own 5% or more of the company. CLSi,t-1 is the percentage of shares closely held by strategic investors. LITi,t-1 is the probability of litigation and is estimated as shown in equation (2).

ED

MA NU

SC

RI P

T

Appendix 2 Corporate Giving, Investor Perceptions and Value Relevance without PHILi,t and INCi,t Panel A Equation (3) Panel B Equation (5) Variables Coefficients Variables Coefficients BVi,t-1 0.002 1/BVi,t-1 0.013* (0.351) (0.008) AEPSi,t-1 0.046** EAR/BVi,t-1 -0.001 (0.023) (0.024) CGi,t 0.083** CGi,t 0.125*** (0.037) (0.036) EAR/BVi,t-1 x CGi,t CFLOWi,t-1 0.008* 0.394*** (0.004) (0.083) CFLOWi,t-1 x CGi,t 0.334*** CFLOWi,t-1 0.082** (0.132) (0.040) Constant 0.023 CFLOWi,t-1 x CGi,t 0.151*** (0.023) (0.050) Constant 0.078 (0.074) R2 adj. R2 adj. 0.612 0.460 Sample size Sample size N=3,435 N=3,435 ***, ** and * indicate statistical significance at the 1%, 5% and 10% level (two-tailed) respectively. The standard error is in parentheses. In Panel A, the dependent variable is Pi,t, which is the stock price at the end of the year. In Panel B, the dependent variable is MBVi,t, which is market to book value of equity. BVi,t-1 is book value of equity. AEPSi,t-1 is earnings per share at the end of the year, minus the cost of equity capital multiplied by book value of equity at the beginning of the year. CGi,t is annual corporate giving scaled by total assets. CFLOWi,t-1 is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. EAR/BVi,t-1 is net income before extraordinary items scaled by book value of equity.

AC

CE

PT

Appendix 3 Corporate Giving and Discretionary Accruals without PHILi,t and INCi,t Variables Coefficients CGi,t -0.111* (0.065) lnFREi,t-1 -0.814*** (0.001) BELi,t-1 -0.001 (0.002) HIGHi,t-1 0.015 (0.010) BELi,t-1 x CGi,t -0.595*** (0.024) HIGHi,t-1 x CGi,t 0.036 (0.026) ICFi,t-1 0.006 (0.008) lnAi,t-1 0.019 (0.017) CFLOWi,t-1 0.007*** (0.002) MBVi,t-1 0.861*** (0.010) CGi,t x lnAi,t-1 -0.0018* (0.001) CGi,t x CFLOWi,t-1 0.002** (0.001) CGi,t x MBVi,t-1 1.240*** (0.028) IDBi,t-1 -0.022 (0.023) NEXECi,t-1 -0.020** (0.010)

36

ACCEPTED MANUSCRIPT -1.009*** (0.036) CLSi,t-1 -0.008 (0.015) Constant 0.089 (0.039) R2 adj. 0.593 Sample size N=3,435 ***, ** and * indicate statistical significance at the 1%, 5% and 10% level (two-tailed) respectively. The standard error is in parentheses. The dependent variable is AAi,t-1, which is abnormal accruals and obtained as shown in equation (10). CGi,t is annual corporate giving scaled by total assets. lnFREi,t-1 is the natural logarithm of fundraising expenses. BELi,t-1 = 1 if earnings minus abnormal accruals, scaled by total assets, is less than zero or less than lagged reported earnings and 0 otherwise. HIGHi,t-1 = 1 if earnings minus abnormal accruals, scaled by total assets, is more than zero or more than lagged reported earnings and 0 otherwise. ICFi,t-1 is the standard deviation of operating income scaled by total assets divided by the standard deviation of operating cash flow scaled by total assets. lnAi,t-1 is the natural logarithm of total assets. CFLOWi,t-1 is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. MBVi,t-1 is market to book value of equity. IDBi,t-1 is the percentage of independent directors sitting on the board of directors. NEXECi,t-1 is the ratio of non-executive to executive directors. BIGAUi,t-1 = 1 for firms that are audited by a big 4 auditor and BIGAUi,t-1 = 0 otherwise. CLSi,t-1 is the percentage of shares closely held by strategic investors.

MA NU

SC

RI P

T

BIGAUi,t-1

AC

CE

PT

ED

Appendix 4 Corporate Giving and Discretionary Accruals: A Comparison with Non-Givers Variables Coefficients dCGi,t -1.075*** (0.016) BELi,t-1 -0.006*** (0.001) HIGHi,t-1 0.002*** (0.0008) BELi,t-1 x dCGi,t -1.164*** (0.105) HIGHi,t-1 x dCGi,t 0.517*** (0.075) ICFi,t-1 0.044*** (0.016) lnAi,t-1 0.010 (0.011) CFLOWi,t-1 0.014 (0.030) MBVi,t-1 1.117*** (0.436) dCGi,t x ICFi,t-1 1.531* (0.810) dCGi,t x lnAi,t-1 -0.010*** (0.002) dCGi,t x CFLOWi,t-1 0.005* (0.003) dCGi,t x MBVi,t-1 0.122 (0.143) IDBi,t-1 -0.019** (0.009) NEXECi,t-1 -0.055* (0.033) BIGAUi,t-1 -0.457*** (0.070) CLSi,t-1 -0.213 (0.390) Constant 0.179 (0.341) R2 adj. 0.531 Sample size N1 =3,435, N0 =7,325 ***, ** and * indicate statistical significance at the 1%, 5% and 10% level (two-tailed) respectively. The standard error is in parentheses. The regression equation is as follows: AAi,t-1 = a0 + a1 dCGi,t + a2 BELi,t-1 + a3 HIGHi,t-1 +

37

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

a4 BELi,t-1 x dCGi,t + a5 HIGHi,t-1 x dCGi,t + a6 ICFi,t-1 + a7 lnAi,t-1 + a8 CFLOWi,t-1 + a9 MBVi,t-1 + a10 dCGi,t x ICFi,t-1 + a11 dCGi,t x lnAi,t-1 + a12 dCGi,t x CFLOWi,t-1 + a13 dCGi,t x MBVi,t-1 + a14 IDBi,t-1 + a15 NEXECi,t-1 + a16 BIGAUi,t-1 + a17 CLSi,t-1 + e. The dependent variable is AAi,t-1, which is abnormal accruals and obtained as shown in equation (10). dCGi,t = 1 for firms that are corporate givers and 0 otherwise. Corporate givers amount to 687 (see Section 4.1), and non-corporate givers amount to 1,465. The sample of non-corporate givers is derived as follows: 2,152 non-financial firms listed on NYSE (see Section 4.1) minus 687 corporate givers. The period of investigation is 2008 to 2012. BELi,t-1 = 1 if earnings minus abnormal accruals, scaled by total assets, is less than zero or less than lagged reported earnings and 0 otherwise. HIGHi,t-1 = 1 if earnings minus abnormal accruals, scaled by total assets, is more than zero or more than lagged reported earnings and 0 otherwise. ICFi,t-1 is the standard deviation of operating income scaled by total assets divided by the standard deviation of operating cash flow scaled by total assets. lnAi,t-1 is the natural logarithm of total assets. CFLOWi,t-1 is operating income before extraordinary items minus depreciation minus interest expense minus preferred cash dividends minus common cash dividends, scaled by total assets. MBVi,t-1 is market to book value of equity. IDBi,t-1 is the percentage of independent directors sitting on the board of directors. NEXECi,t-1 is the ratio of non-executive to executive directors. BIGAUi,t-1 = 1 for firms that are audited by a big 4 auditor and BIGAUi,t-1 = 0 otherwise. CLSi,t-1 is the percentage of shares closely held by strategic investors.

38

ACCEPTED MANUSCRIPT Highlights

T

Structured philanthropy and in-kind contributions encourage corporate giving.

RI P

Leverage and corporate governance affect corporate giving downwards.

SC

Corporate giving is value relevant and positively influences investor perceptions.

MA NU

Corporate giving is positively to analyst coverage.

AC

CE

PT

ED

Charitable firms tend to engage less in earnings manipulation.

39