How and Why Do Corporations Combine Lobbying with Social Responsibility?

Transcription

1 How and Why Do Corporations Combine Lobbying with Social Responsibility? Patrick Bernhagen, University of Stuttgart and University of Aberdeen Natalka Patsiurko, Concordia University This version: 29 August 2014 This is a very first draft. Please do not cite without the authors explicit permission. Abstract Lobbying by business interests is of major public and academic concern. At the same time, corporations are increasingly engaging in various activities related to corporate social responsibility (CSR) and citizenship. While these two modes of non market activity are fundamentally political in nature, business actors as well as researchers are frequently uncertain about how they (should) relate to each other: How do corporations combine more overtly self interested corporate political activity (CPA) and with expectations of corporate social responsibility and citizenship? To investigate these questions, we analyze a new dataset of almost 1,000 corporations from Germany, the UK and the US that contains information on their political and social activities. We develop and test empirical models of CPA and CSR to evaluate expectations that these two modes of behaviour can be explained by the same behavioural logic. Our findings show that CPA and CSR can be explained by the same theoretical arguments and empirical models. Moreover, engaging in CPA significantly increases the likelihood of a corporation to employ CSR related strategies. Paper for presentation at the ECPR General Conference, University of Glasgow 3 6 September The research for this paper has been supported by the Economic and Social Research Council (grant ID ES/I036974/1).

2 How and Why Do Corporations Combine Lobbying with Social Responsibility? It is difficult to remember a time when questions about the role that corporations play in politics have been a more passionately debated subject than it is today. It has also arguably never been more important than today for political scientists and democratic theorists to understand the contributions that corporations make to creating political problems as well as solutions. Five years after the global financial and economic crisis began, and after a decade of high profile corporate scandals in multiple countries, governments, organized groups and ordinary citizens are questioning the corporation s proper place in society. In advanced industrial democracies, business participation is the largest component of interest group activity, whether measured by the number of participants or the resources committed to lobbying (Baumgartner and Leech 2001; Grant 1993; Greenwood 2007). But to date political scientists have devoted insufficient attention to the role that corporations play in politics (Wilks 2013). One particular area of neglect concerns the way in which this role varies across political venues and types of activity. While citizen participation tends to be territorially limited, many business corporations have the resources and incentives to take up multiple activities in various national and international arenas. Traditionally, capitalist democracies relied on the authority of the state to regulate corporate actors, but in recent years transnational policymakers, a host of non governmental groups and alternative governance modes beyond the state have emerged. Globalization and the evolution of multi level systems of governance thus pose new challenges and opportunities for corporations and those actors seeking to influence their behaviour. In order to increase the levels of accountability and transparency necessary for democratic politics, policymakers and the public need information about the kind of organizations that have interests in a policy and the access they have to elected and appointed officials. Policymakers and civil society groups interested in increasing accountability in public life need empirical analyses of corporate social responsibility (CSR) and corporate political activity (CPA) to evaluate the efficacy of various programmes for regulating and monitoring corporations political participation as well the effectiveness of social and environmental responsibility schemes. Attempting to provide such analysis involves tackling two different frontiers: First, as global markets continue to integrate and political decision making expands to inter and supranational levels, corporations face incentives to develop their non market strategies accordingly to engage these new frontiers of political and social engagement. Secondly, political scientists have only periodically contributed to the crucial debates about the corporation s role in society with studies that seek to uncover how corporations participate in 2

3 politics and policymaking processes. As the literature to date has not adequately addressed the variety of forms corporate political participation can take, a new frontier in academic research opens up. In addition to lobbying and donating to political parties, corporations, in increasing numbers, are participating in self regulatory schemes to demonstrate their commitment to corporate social responsibility (CSR). And while we know that some corporations are active in several types of nonmarket strategy at once, we do not know whether these are employed independently or whether they are linked strategically. As some commentators fear, corporations might use CSR strategically to avert stricter regulation and real accountability (Vogel 2010). If this is the case, then any achievements of global governance goals resulting from CSR related activities may be undone by the setbacks that result from anti regulatory lobbying facilitated by the very same schemes. This question is yet to be addressed by systematic research. Explaining how and why actors participate in politics is a core area of political research, but the field remains surprisingly under informed about corporate political behaviour despite widely shared concerns that corporations exert considerable influence over public policy. Most of the work on corporate political behaviour has concentrated on single countries and focussed on one mode of engagement (e.g. lobbying, campaign contributions, or corporate social responsibility) at a time. As a result, we have little understanding of how corporate behaviour varies across different political arenas and modes of activity. Against this backdrop, the present paper seeks to improve our understanding of how and why corporations participate in politics by analyzing the non market behaviour of 953 transnational corporations across different types of activities lobbying and corporate social responsibility and venues EU and UN. To this end, we have compiled a new dataset containing information on corporation size, country of headquarter, and industry, as well as on corporations lobbying and corporate social responsibility (CSR) activities at these two different political venues. In addition to gaining insight into the extent and patterns of corporate political participation, our analysis aims to contribute to theoretical debates on corporate behaviour in political science and beyond. In the remainder of the paper, we proceed as follows. The next section briefly reviews the literatures on CPA and CSR and develops our argument about the relationship between these two forms of non market activity. After that, we describe the data we collected to evaluate out argument before presenting results if preliminary empirical analysis. We conclude by discussing our findings in the light of the extant literature on corporate non market activity. 3

4 Explaining CPA and CSR Corporations engage in non market activities in a variety of ways, often with the aim of influencing public policy and preventing adverse regulation (Grier et al. 1994, Stigler 1971). As they engage politically, corporations are aware that their public reputations and the trust of key external actors in the political arena and civil society are important political assets. Managing these assets is a key part of corporate non market strategies. While research on corporate political activity has mainly focused on corporate lobbying of policy makers and financial contributions to electoral campaigns, there is growing recognition that corporations are expanding their repertoire of political activity as part of an integrated non market strategy (Baron 2010, Rehbein et al. 2011, Schuler et al. 2002). In this context, the legitimacy that a corporation can gain by participating in public private voluntary initiatives can reduce the threat of adverse consumer, public or political activity (Meznar and Nigh 1995). Consequently, scholars have begun to examine the political dimensions of what is generally referred to as corporate social responsibility (CSR) (Hillman and Hitt 1999, Matten and Crane 2005, Waddock 2008), including the political and institutional factors that shape corporate decisions to participate in CSR related activity and voluntary codes (Brammer and Pavelin 2006, Doh and Guay, 2006). Of central concern is the considerable variation in participation across different countries, economic sectors and types of corporations (Kollman and Prakash 2001, Bennie et al. 2007, Perez Batres et al. 2011). For example, while in the year 2010 only 2.4 corporations per 100,000 inhabitants of Egypt have signed up to the UN Global Compact, 24 corporations from South Korea have that is ten times as many. Researchers have tried to explain this variation referring to a variety of factors including corporations individual characteristics, differences in the mobilization of normative biases within their home countries, the behaviour and strategies of countervailing interests, and the wider institutional environment in which corporations exist and operate. We contribute to this literature by analyzing the choice of non market activity employed by corporations. Here, the options range from the more instrumental and specific (e.g. lobbying) to the less specific and symbolic, e.g. CSR (Edelman 1964; Nye 2004). The questions are: How do corporations activities in one political arena affect their strategies in others? Is there a subset of corporations and sectors active in multiple arenas? And do the same factors explain both the more instrumental corporate activities such as lobbying and their less instrumental activities such as corporate social responsibility? 4

5 To address these questions, we draw on conventional models of corporate political activity (Grier, Munger and Roberts 1994; Hansen, Mitchell and Drope 2005) as well as on management studies on CSR (Crane et al 2008; Josselin and Wallace 2001; Kolk 2003). Our strategy is to start with a simple, profitseeking model of the corporation. We expect several factors to structure corporations incentives for participating in the various political activities: corporation characteristics (size, ownership), sector characteristics such as the degree of competition), and home country characteristics including the politico economic system and the normative and cultural environment). Now, a set of explanatory factors can be outlined. A corporation s size reflects its resources and market power as well as what it has at stake in economic and political conflict (Grier et al. 1994). As larger corporations have more to gain from political activity than smaller ones, at least in absolute terms, they will often ignore the free rider incentives of others (Olson 1965). Larger corporations often also have increased reputational incentives for participating in political activity as they have to protect publically visible brand names. Indeed, individual corporation size has consistently been an important determinant not only of political activity but also of CSR related activities and voluntary disclosure (Hillman et al. 2004: 839, Meek et al. 1995). Similarly large corporations are more likely to participate in global initiatives. Research on the non market behaviour of the Forbes Global 2000 corporations, for example, shows that larger corporations are significantly more likely to sign the UN Global Compact (Bennie et al. 2007). Differences in corporate governance systems might also affect the decisions that corporations make about participating in non market activities. As Kolk and van Tulder (2005) argue for CSR, the outsider system of corporate governance dominant in the United States combines diffuse shareholding with a prominent role of the CEO. Together with its high propensity for liability and class action suits, this system, they reason, might encourage socially responsible investment (Kolk and van Tulder 2005: 8). The European and Japanese systems of corporate governance, by contrast, are insider systems whose twotier board structure and less prominent CEO status might combine to use codes of conduct more as an internal control (rule setting) instrument (ibid.). However, no empirical evidence exists to date to support this expectation. What is supported, however, is the view that a corporation s involvement in political activity can be shaped by the personal preferences and ideals of management (Ozer 2010). 5

6 H1: Corporate participation in lobbying and participation in CSR activities are determined by the same variables. Following Richter (2011), we argue further that corporations CSR behaviour works as an economic complement to its political activity rather than a substitute. The argument is quite straightforward: Credibility and legitimacy are key assets underpinning any CPA strategy, such as lobbying. Displaying a credible commitment to CSR contributes to the production and maintenance of these assets. This implies that corporations for whom lobbying is an important element in their non market action repertoire have increased incentives to engage in CSR related activities compared with corporations for whom lobbying is less central. Thus, we expect that, H2: Increased propensity to engage in CSR is associated with an increased likelihood of establishing lobby infrastructure, ceteris paribus. The next section outlines how we intend to address these issue with the help of improved empirical data. Data, measurement and estimation strategy In contrast to individual participation by citizens, where there is an extensive range of cross national datasets from election studies and opinion surveys, there are no comparable datasets for corporationlevel political action. To track different types of corporate political engagement in multiple national, supra, and transnational policy arenas, we have compiled information on the political activities of 953 corporations from Germany, the UK and the US three countries representing important politicoeconomic system types. Corporations demographic and economic data are taken from Standard & Poor s Capital IQ database, which contains almost 60,000 public and more than 1.7m private companies from all over the world. From this source we have sampled the 300 largest corporations from reach of the three countries. For this purpose, we have created a size ranking based on revenues, assets and employee numbers. For the US, we have added an additional 51 private corporations, as the top three hundred US forms are disproportionately dominated by publically traded corporations. To the 300 UK corporations, two based on the Channel Islands were added. 6

7 The Capital IQ data contain corporation level information on size, sales, sectoral activity, and type of ownership. In the analysis that follows, size is measured by revenues in million US Dollars made during the last twelve months before our point of measurement in late Profit is gross profit revenues in million US Dollars realized during the same time period. Private corporations are those private equity and family owned corporations that are not indexed in any stock market. We expect these three variables to be associated with non market activities, and to be so in the same way for CPA and CSR. To this base of observations we have added information on corporate political action. Specifically, we have collected data on the political activities of these 953 corporations from public lists and private directories. Lobbying data for the EU has been collected through corporation level searches of the 2013 European Public Affairs Directory (Harris 2013). A corporation is coded as engaging in EU level CPA if it entertains a public affairs representative with an office in Brussels (Brussels office). Setting up a European Affairs post is costly while being potentially useful for targeting all three major policy making bodies: the Commission, the Council of Ministers and the EP. Equipping this representative with his or her own office in Brussels is more costly still, but a Brussels base is useful for firms seeking to play a role in European public affairs. CSR is measured by signing on to the human rights, environmental and good governance principles that constitute the UN Global Compact (UNGC).The dataset thus provides us with binary measures of corporate non market activity for in the EU and the UN (coded 1 if the corporation engages in the activity, 0 otherwise). Controlling for context at the level of country and industry In addition to these firm level variables, a host of contextual factors must be controlled for that may affect both the propensity to act politically as well as the relationship between the aforementioned firmlevel variables and corporate non market strategies. Many of these concern institutions understood here broadly as the rules of the game that structure and constrain actors behaviour (North 1990: 97). These rules can be formal (legally binding rules and regulations) or informal (soft law codes of conduct, norms of behaviour, and conventions). Nation states are the foundations of the most important institutions in the international system, and even globally active corporations operate on the basis of nationally distinct politico economic systems. Thus, general differences in national styles of capitalism as captured, e.g., by the distinction between liberal market economies and co coordinated market economies (Hall and Soskice 2001). In liberal market economies, like the UK or US, corporations coordinate their activities primarily via hierarchies and competitive market arrangements, while in 7

8 coordinated market economies, such as Germany, corporations depend more heavily on non market relationships to coordinate their relationships with other actors (Hall and Soskice 2001: 8). More accustomed to taking responsibility for their employees and the concerns of external actors, corporations from coordinated market economies in continental Europe might be expected to be more likely to agree to long term commitments to the principles of human rights, the society and the environment (Janney et al. 2009). 1 To the extent that political institutions offer different political opportunity structures (Kitschelt 1986), the propensity to participate politically will differ across political arenas (Marks and McAdam 1996; Naoi and Krauss 2009; North 1997). While a number of parallels between the US and EU political systems make comparative analysis across these different systems meaningful (Menon 2006), procedural differences are likely to elicit different lobbying patters. Decisions to participate further take place within particular institutional and normative contexts (cf. Huckfeldt and Sprague 1993). We expect the relevance of particular neighbourhood cues and signals to inform a corporation s choice of political activity. Third, to the extent that organizations respond to what is considered appropriate in their social contexts (March and Olson 1989), we take into account the influence that the social norms and cultural values of corporations home and host countries have on their behaviour. Existing analyses suggest that global corporations react to the pressure of countervailing powers, such as NGOs and labour unions, and adapt to the normative context of their home countries: corporations located in countries with strong environmental movements are more likely to participate in the UN Global Compact (Bennie et al. 2007). Furthermore, the non market behaviour of global corporations is partly shaped by national regulatory regimes. Countries have different rules on lobbying and financial contributions as well as different requirements for corporate reporting in areas of social or environmental responsibility. Increasingly, such requirements are legally mandated, placing similar demands on corporations as voluntary schemes in these areas do. In Denmark, the Financial Statement Act was amended in 2009 to require reporting on corporate social responsibility. France has had a legal requirement since 1977 for corporations with more than 300 employees to publish a social review (KPMG et al. 2010). More generally research on the 1 Like the argument about corporate governance, the role of nationally distinct types of capitalism has not yet been researched empirically in great depth. 8

9 uptake of environmental management systems has shown that the regulatory environment affects participation, although researchers disagree on the nature of this effect (Prakash and Potoski 2006: 142 4; Neumayer and Perkins 2004; (Kollman and Prakash 2001). Table 1 shows the levels of corporate activity for firms headquartered in each of the three countries in our dataset. As can be seen from Table 1, about 8 percent of all corporations in our data entertain a lobbying office in Brussels. Over 14 percent of corporations have signed up to the UNGC. German corporations are most active in both modes of engagement. Perhaps surprisingly, the second most active country in terms of EU level lobbying is the US; while corporations from this country are less likely than their German and UK counterparts when it comes to UNGC membership. Table 1. Percentages of corporations participating in non market activities, by country Germany UK US All Brussels Office UNGC N Corporations involvement in politics and CSR is also likely to be shaped by the nature of their sectoral activity. Corporations in extractive sectors are exposed to a higher risk of conflict with external actors, especially in the areas of environment and human rights. Natural resources such as oil, gas and minerals are often fixed and located in politically as well as geographically difficult environments. For global corporations active in these locations, reputation building, political communication and engagement to safeguard their investment become important political activities. Hence, these resource cursed corporations seek opportunities to enhance their reputation and take counter measures to increase their public legitimacy (Bennie et al. 2007) as well as lobbying policymakers. At the industry level, corporations political engagement is likely to be shaped by their involvement in transnational business networks (Coen and Grant 2001; Perkins and Neumayer 2010) and by the extent to which they can rely on trade or industry associations to pursue their political interests (Streeck et al. 2006). In our data, we distinguish by industry according to the divisions of the Standard Industrial Classification (SIC). The distribution of corporations in our data across these divisions is shown in Table 2. Summary statistics of all other variables are reported in Table A 1 in the appendix. 9

10 Table 2. Representation of industrial sectors in the data Sector (SIC division) N Percent Agriculture, Forestry and Fishing Mining, Oil and Gas Construction Manufacturing Transport, Communication and Public Utilities Wholesale Trade Retail Trade Finance, Insurance and Real Estate Services Conglomerates and multi sector holdings Total Results As these are new data, we approach the empirical evaluation of our hypotheses from a descriptive and bivariate angle first. As can be seen from Table 3, the percentage of corporations, just over 14 percent in the sample, rises steeply to almost half of all firms in the data when the focus is on firms with active political presence in Brussels. Thus, CSR and CPA tend to go me hand in hand. The association is highly statistically significant. Table 3. Percentages of corporations participating in the UNGC, by Brussels office Brussels Office UNGC No Yes All Non participant Participant Total Pearson s Chi square = 85.69, p<

11 The bivariate correlation coefficient between these two measures of non market activity is φ = 0.30 (see Table 4). This is approximately the strength of the relationship between lobbying and CSR observed among UF corporations. The association is weaker for British firma and stronger for German firms. Table 4. Bivariate correlations between non market activities, by country Germany UK US All Phi 0.44* 0.13* 0.28* 0.30* N Note: * p<0.05, To gauge the extent to which the same variables serve to predict lobbying and CSR activities, we estimate generalized linear models with fixed effects and random intercepts using a logit link function. There are random intercepts at two levels countries and industrial sectors. As these levels are not nested, we estimate crossed effects. Through random intercepts at country and industry level, most of the factors discussed under context in the previous section are controlled for in the background; even we do not explicitly estimate their effects here. As can be seen from Table 5, the fixed effects estimates of the coefficients for ownership type, revenues and profit are quite similar across the two modes of non market activity. The ownership type does not affect the likelihood of either lobbying or signing up the the UNGC. Size as measured by revenues is a significant predictor of both CPA and CSR. So is profit, although the effect is small even when taking into account that the square root of this variable has been used for better fit. Overall, these estimates suggest that the amen factors account for CSR that also serve to explain variation among corporations in respect of lobbying. 11

12 Table 5. Multilevel mixed effects logistic regression for CPA and CSR behaviour Brussels Office UNGC Brussels Office UNGC Private (0.58) (0.20) (0.67) (0.20) Ln Revenues 1.99*** 1.79*** 1.77** 1.70*** (0.28) (0.20) (0.33) (0.19) Profit (square root) 1.01** 1.01** 1.01* 1.01* (0.00) (0.00) (0.00) (0.00) UNGC 3.62*** (1.17) Brussels Office 3.10*** (0.99) Constant 0.00*** 0.00*** 0.00*** 0.00*** (0.00) (0.00) (0.00) (0.00) Country level (0.61) (2.70) (0.25) (2.12) Sector level (1.90) (0.21) (1.05) (0.20) Log likelihood Wald 90.80*** *** 90.86*** *** Note: N=953; * p<0.05, ** p<0.01, *** p< Cell entries are odds ratios. 12

13 In columns 4 and 5 of Table 5, UNGC participation has been included as a predictor the CPA measure Brussels office (column 4), and vice versa (column 5). The large and highly significant coefficients for the non market measures confirm that the finding from the bivariate analysis two modes of participation go together even when other variables are controlled for. However, because it is possible that firms that are strong on CSR are also more likely to develop lobbying infrastructure, these estimates may be marred by endogeneity of the political action measure in the right hand side of the equation. Therefore, in order to assess Hypothesis 2, we estimate an instrumental variable probit model in which the endogenous Brussels office is instrumented by an exogenous size measure, total assets. This measure is unrelated to UNGC but significantly associated with Brussels office. Thus, we are able to estimate the effect of having a Brussels office on the likelihood of joining the UNGC after ensuring that this predictor is no longer correlated with the error term of the empirical model. These estimates confirm the significant association between the two non market action variables. After controlling for a number of firm level predictors as well as country dummies and clustering of standard errors on industry, corporations that lobby in Brussels are twice as likely to join the UNGC as those that are not engaged in this form of CPA. 13

14 Table 6. Instrumental variable probit regression for CSR behaviour Probit coefficients (2 nd stage) Brussels Office 2.07*** (0.45) Private 0.36 (0.26) Profit (square root) 0.00* (0.00) United Kingdom 0.11 (0.15) United States 0.78*** (0.22) Constant 1.19*** (0.18) Arc hyperbolic tangent of ρ 0.35 (0.21) ln σ 1.38*** (0.08) LL Wald P Note: N=953; * p<0.05, ** p<0.01, *** p<0.001; standard errors in parentheses corrected for clustering on industrial sector; instrument: Total Assets 14

15 Conclusions We have developed and tested empirical models of CPA and CSR to evaluate expectations that these two modes of behaviour can be explained by the same behavioural logic. Analyzing a new dataset of almost 1,000 corporations from Germany, the UK and the US, we have shown that CPA and CSR can be explained by the same theoretical arguments and empirical models. Moreover, engaging in CPA significantly increases the likelihood of a corporation to employ CSR related strategies. The findings provide reasons for viewing CSR activities as part of corporations political action repertoire, where they serve as a buffer that smoothens relationships with public decision makers and alleviates any fallout from negative publicity related to lobbying. 15

16 References Baron, D. (2010) Business and Its Environment, 6 th ed., Englewood Cliffs, NJ: Prentice Hall. Bennie, L., Bernhagen, P., Mitchell, N. J. (2007) The Logic of Transnational Action: The Good Corporation and the Global Compact, Political Studies, 55(4): Brammer, S. J., Pavelin, S. (2006) Corporate Reputation and Social Performance: The Importance of Fit. Journal of Management Studies, 43(3): Coen, D. and W. Grant (2001) Corporate political strategy and global policy: a case study of the Transatlantic Business Dialogue, European Business Journal 13 (1): Crane, A. et al. (2008) The Oxford Handbook of Corporate Social Responsibility. Oxford: Oxford UP. Cutler, C., V. Hauler und T. Porter, eds. (1999) Private authority and international affairs. New York: SUNY Press. Doh, J. P., Guay, T. R. (2006) Corporate Social Responsibility, Public Policy, and NGO Activism in Europe and the United States: An Institutional Stakeholder Perspective, Journal of Management Studies, 43(1): Grier, K. B., Munger, M. C., Roberts, B. E. (1994) The Determinants of Industry Political Activity, , The American Political Science Review, 88(4): Hall, P. A., Soskice, D. W. (2001) Varieties of Capitalism: The Institutional Foundations of Comparative Advantage, Oxford: Oxford University Press. Harris, T. (2013) European Union and Public Affairs directory London: Dod s Parliamentary Communications. Hillman, A. J., Hitt, M. A. (1999) Corporate Political Strategy Formulation: A Model of Approach, Participation, and Strategy Decisions, The Academy of Management Review, 24(4): Hillman, A. J., Keim, G. D., Schuler, D. (2004) Corporate Political Activity: A Review and Research Agenda, Journal of Management, 30(6): Janney, J. J., Dess, G., Forlani, V. (2009) Glass Houses? Market Reactions to Corporations Joining the UN Global Compact, Journal of Business Ethics, 90(3): Josselin, D. and Wallace, W. (2001) Non state Actors in World Politics. London: Palgrave. Kitschelt, H.P. (1986) Political Opportunity Structures and Political Protest: Anti Nuclear Movements in Four Democracies, British Journal of Political Science 16(1): Kolk, A. (2003) Trends in Sustainability Reporting by the Fortune Global 250. Business Strategy and the Environment, 12,

17 Kolk, A., van Tulder, R. (2005) Setting New Global Rules? TNCs and Codes of Conduct, Transnational Corporations, 14(3): Kollman, K., Prakash, A. (2001) Green by Choice? Cross national Variations in Corporations Responses to EMS based Environmental Regimes. World Politics, 51(3): Marks, G. and D. McAdam (1996) Social Movements and the Changing Structure of Opportunity in the European Union. West European Politics 19 (2): Matten, D., Crane, A. (2005) Corporate Citizenship: Towards an Extended Theoretical Conceptualization, Academy of Management Review, 30(1): Meek, G. K., Roberts, C. B., Gray, S. J. (1995) Factors Influencing Voluntary Annual Report Disclosures by US, UK and Continental European Multinational Corporations, Journal of International Business Studies, 26(3): Menon, A. (2006), The Limits of Comparative Politics: International Relations in the European Union, in A. Menon and M. Schain (eds.), Comparative Federalism (Oxford: Oxford UP). Meznar, M. B., Nigh, D. (1995) Buffer or bridge? Environmental and Organizational Determinants of Public Affairs Activities in American Corporations, Academy of Management Journal, 38(4): Naoi, M., and E. Krauss (2009) Who Lobbies Whom? Special Interest Politics under Alternative Electoral Systems, American Journal of Political Science 53(4): Neumayer, E., Perkins, R. (2004) What Explains the Uneven Take up of ISO at the Global Level? A Panel Data Analysis, Environment and Planning, A 36(5): North, D. C. (1990) Institutions, Institutional Change, and Economic Performance, New York: Cambridge University Press. Olson, M. (1965/1971) The Logic of Collective Action: Public Goods and the Theory of Groups, Cambridge, MA: Harvard University Press. Ozer, M. (2010) Top Management Teams and Corporate Political Activity: Do Top Management Teams have Influence on Corporate Political Activity?, Journal of Business Research, 63(11): Perez Batres, L. A., Miller, V. V., Pisani, M. J. (2011) Institutionalizing Sustainability: An Empirical Study of Corporate Registration and Commitment to the United Nations Global Compact Guidelines, Journal of Cleaner Production, 19(8):

18 Perkins, R., Neumayer, E. (2010) Geographic Variations in the Early Diffusion of Corporate Voluntary Standards: Comparing ISO14001 and the Global Compact, Environment and Planning C: Government Policy, 42(2): Prakash, A. and Potoski, M. (2006) The Voluntary Environmentalists: Green Clubs, ISO 14001, and Voluntary Environmental Regulations. Cambridge: Cambridge University Press., Rehbein, K.A. et al. (2009) Corporate responses to Shareholder activism: An institutional perspective, Academy of Management Proceedings, Chicago, IL. Schuler, D. A., Rehbein, K., Cramer, R. D. (2002) Pursuing Strategic Advantage through Political Means: A Multivariate Approach, Academy of Management Journal, 45(4): Streeck, W., J. Grote, V. Schneider and J. Visser (eds) (2006) Governing Interests. London: Routledge. Vogel, D. (2010) The Private Regulation of Global Corporate Conduct, Business and Society, 49(1): Waddock, S. (2008) Building a New Institutional Infrastructure for Corporate Responsibility, The Academy of Management Perspectives, 22(3):

19 Appendix Table A 1. Summary statistics of all variables except sector Variable N Mean Std. Dev. Min Max Brussels Office UNGC Private Revenues Profit Assets Table A 2. First stage estimates First stage Private 0.03 (0.02) Profit (square root) 0.00*** (0.00) United Kingdom 0.06 (0.04) United States 0.13** (0.04) Total Assets 0.00** (0.00) Constant 0.03 (0.04) 19