Assessing the Legitimacy of Corporate Political Activity: Uber and the Quest for Responsible Innovation

Building on literature in political CSR and corporate political activity (CPA) as well as responsible innovation and responsible lobbying, we introduce a framework to assess the legitimacy status of corporate political activity. We focus on the fact that companies frequently face sharp regulatory backlash after penetrating markets with their innovations. In response to regulatory backlash, big tech companies often employ an arsenal of corporate political activities to (re-)shape national and local regulatory environments, which raises the important questions about the legitimacy of CPA tactics that we address. To develop and apply this framework, we briefly survey and then assess Uber’s corporate political activities in its 2015 New York City political campaign.


Introduction
The innovations in technology that have transformed society at a blistering pace in the twenty-first century-from smartphones to social media-epitomize the opportunities and threats business creates for people and planet. Information is being generated and shared in previously unthinkable ways to promote valuable human ends-and make extraordinary fortunes. However, the same technologies are also being used to produce government-driven surveillance systems (Zuboff, 2019), social scoring systems (Liang et al., 2018), and voter manipulation (Bradshaw & Howard, 2018) while introducing problematic bias (Martin, 2021) and causing widespread addiction (Abroms, 2019;Bhargava & Velasquez, 2020) and untenable and deadly greenhouse emissions (Belkhir & Elmeligi, 2018;Schwartz, 2021) to highlight just a few issues that recent innovations have spawned. Led by scientists and technologists, scholars have responded to the mixed track record of new technologies with calls for responsible innovation (Von Schomberg, 2012).
In leading articles, Owen et al. (2013) and Stilgoe et al. (2013) reformulate sustainability concepts (Brundtland, 1987) to conceive of "[r]esponsible innovation [as] taking care of the future through collective stewardship of science and innovation in the present" (Stilgoe et al., , p. 1570. The reception of the project of responsible innovation (RI) from the natural sciences and engineering to organization theory and management science has been pursued most systematically by Voegtlin and Scherer in a series of articles (Patzer et al., 2018;Voegtlin & Scherer, 2017;Voegtlin et al., 2018) that articulate a "responsible governance" framework. This framework appeals to the political CSR (PCSR) approach pioneered by Scherer andPalazzo (2007, 2011;Scherer et al., 2016), spelled out to address problematic innovation with a call to do good. As in the contributions that popularized PCSR, the articulation of the responsible innovation framework appeals to "open systems" (Filatotchev et al., 2020) forms of 1 3 "deliberative democracy" (Scherer & Voegtlin, 2020, p. 192) predicated upon voluntary participation in multi-stakeholder initiatives formed for deliberation and soft law norm-making (de los Reyes et al., 2017).
While the question of responsible innovation in general Stilgoe et al., 2013) and how to govern responsible innovation in particular Voegtlin & Scherer, 2017) has been addressed by scientists and organization theorists, the role and relevance of corporate political activities (CPAs) designed to shape the hard law regulatory environment has yet to be directly explored. In this article, we focus on the fact that companies often face sharp regulatory backlash after penetrating new markets with their innovations. In response to regulatory backlash (sometimes also preemptively), big tech companies tend to employ a toolkit of CPA tactics to (re-)shape national and local regulatory environments (Uzunca et al., 2018). Amazon, Apple, Facebook, and Google, to name four prominent "big tech" companies, have already eclipsed the big oil and big tobacco industries in terms of money spent on CPAs-Amazon and Facebook reportedly spent nearly double Exxon and Philip Morris on lobbying in 2020-and these four big tech firms spent $124 million on campaign contributions last year (Chung, 2021). These companies have been called "regulatory entrepreneurs" for "pursuing a line of business in which changing the law is a significant part of the business plan" (Pollman & Barry, 2016, p. 383); the innovative business models with which they infiltrate markets simply are not contemplated under existing regulations when launched, creating hard law regulatory voids that tech firms naturally want to fight to keep from being closed down to business. We seek to build theory on the premise that political engagement to shape hard law is a necessary part of doing business for big tech companies so as to then explore how they can push the law to accommodate innovation responsibly-innovation that is responsible in part because the company sustains moral legitimacy in shaping law. 1 Thus, our core research questions in this article are as follows: what makes CPA tactics morally (il) legitimate, why, and under what conditions? In order to answer these questions, we will rely on the principles of PCSR we expound below and our operationalization thereof to assess the moral legitimacy of the CPA tactics deployed by companies in general and big tech companies in particular. To tease out and apply theoretical extensions, we rely upon the focal case of Uber in New York City in 2015 seeking to prevent a law capping rideshare vehicles in the city from coming into effect.
With this paper, we wish to contribute to three discussions. First, we aim to contribute to literature of responsible innovation by focusing on the undertheorized role of CPAs in the efforts of innovative companies to survive in their regulatory environments, especially in response to the theoretically neglected yet empirically predictable pattern of regulatory backlash that leads companies to answer regulatory reining-in with aggressive counterattacks. Second, we contribute to the CPA literature by asking and systematically answering a question this literature has so far ignored: What makes CPA tactics morally (il)legitimate? CPA scholars have contributed to the management field's understanding of how companies design CPA tactics as means to the end of "produc[ing] public policy outcomes that are favorable to the firm's continued economic survival and success" (Keim & Baysinger, 1988, p. 171). Under the general rubric of "corporate political activities" (Hillman et al., 2004) and permutations like "corporate political activism" (Sethi, 1982), "non-market strategy" (Baron, 1995), and "corporate political strategy" (Hillman & Hitt, 1999), scholars writing from this perspective have shed light on why, when, and how effectively companies use tactics like lobbying, campaign finance, coalition building, grassroots campaigns, and mobilization for legislation through voter initiatives (e.g., Puck et al., 2018;Ridge et al., 2019). However, CPA scholars so far do not provide criteria for-and much less a framework for assessing-the legitimacy of these tactics as deployed. While business ethics and PCSR scholars are concerned with the moral legitimacy of corporate responsibility, the moral legitimacy status of CPA tactics in the management literature is murky and merits study, notwithstanding the important contribution by , 2017 to conceptualize the possibility of responsible lobbying and other CPA tactics that can be pursued legitimately. Here, in the area of business ethics and PCSR, we see our third contribution. We aim to develop and extend the project of , 2017 by providing a deliberative and democratic 1 Legitimacy is not an undisputed term (see Bitektine & Haack, 2015). Differing from the monological tradition in business ethics (e.g., Integrative Social Contracts Theory) that centers moral legitimacy around compliance with so-called hypernorms, PCSR scholars have tended to theorize about legitimacy in constructivist terms (Heugens & Scherer, 2010), pp. 647-648; e.g., Scherer et al., 2013: cf. Scholz et al., 2019. According to these scholars, companies need to engage in discourse with their stakeholders to maintain their moral legitimacy. While moral legitimacy then rests on a social construction, there are a few features that can be objectively examined. The PCSR literature has stressed that the participation of corporations in public will formation discourses is sufficient for moral legitimacy if and only if these engagements satisfy certain rules of discourse (Sect. 4). In brief, the question of (input) legitimacy depends upon who talks to whom and how. By introducing the more fine-grained dimensions of (stakeholder) (1) Inclusion and (2 Modes of communication and decision-making, we suggest that our framework for the assessment of CPA tactics asks exactly these objective questions and thus helps to further analyze the moral legitimacy status of individual CPA tactics. Footnote 1 (continued) account of how companies can deploy CPAs responsibly, answering  call for further research to "test whether these nonmarket strategies comply with the normative demands of political CSR: discourse, transparency, and accountability" , p. 427, citing Scherer & Palazzo, 2007.
In summary, the fundamental research aim and contribution we seek to make in this article is to develop and operationalize a legitimacy assessment framework for CPA tactics that builds upon and extends deliberative approaches to corporate responsibility, including political CSR and corporate citizenship.
Since CPA tactics are often used by big tech regulatory entrepreneurs (Pollman & Barry, 2016) that race to establish cognitive legitimacy around their innovation by aggressively growing market share before hard law regulation catches up (Garud et al., 2020), we develop theory by looking at the tactics used by Uber in response to the New York City mayor's 2015 effort to cap rideshare services' meteoric growth by legislating through the city council.
We proceed as follows: After a more extensive review of the relevant literature on CPAs, responsible innovation, and its PCSR antecedents in the first section, "Corporate Political Activities in the Context of Responsible Innovation", we address the relation of our research question to the approaches so far taken by the CPA and PCSR literatures in the second section, "Letting Corporate Activities Speak for Themselves". Next, in "Uber in New York City", we briefly survey Uber's corporate political activities in the 2015 New York City episode to identify three key CPA tactics that we wish to evaluate with the theory we build in this article. In "Assessing the Legitimacy of CPA Tactics", we operationalize a framework for assessing the legitimacy of CPAs by interpolating the principles of PCSR. In order to do so, we will in part build upon Fung's (2006) institutional design framework for public participation and then test the framework by evaluating Uber's CPA tactics in the subsequent section. We then discuss our contributions and, in concluding, reflect on the implications for our framework of an important objection that recently charged PCSR with "reflect[ing] both a triumph of neoliberal corporate power and a harbinger of democracy's demise" (Rhodes & Fleming, 2020, p. 943).

Corporate Political Activities in the Context of Responsible Innovation
In studying CPAs, management scholars have assumed that the point of CPAs-their excellence as a managerial craftcomes from "produc[ing] public policy outcomes that are favorable to the firm's continued economic survival and success" (Keim & Baysinger, 1988, p. 171). While recognizing that "questions remain as to when political strategies of business firms should be considered legitimate versus when they should be considered dangerous to society and democracy" (Scherer et al., 2009, p. 328), PCSR scholars have concurred in the assessment that CPAs are pursued for instrumental reasons rooted in private interests (Rasche, 2015;Scherer, 2017, p. 388;Scherer et al., 2006Scherer et al., , p. 511, 2009Scherer et al., , p. 330, 2014Scherer, et al., 2013, p. 264;Scherer & Palazzo, 2008, p. 421, 2011Scherer et al., 2016, p. 274). CPA tactics like indirect lobbying through advertising campaigns to shape public opinion, for example, are understood by Scherer et al. (2013) as examples of a "strategic manipulation strategy... where corporations actively influence social expectations by swaying or even manipulating the perceptions of key actors or policy makers in their environment" (p. 263).
Neither CPA tactics nor the instrumental pursuit of corporate self-interest can be found in the "normative commitments to democratization" (Stilgoe et al., , p. 1577) that define the "Responsible Research and Innovation" framework that Voegtlin and Scherer Voegtlin et al., 20182017; translate from the scientific to the management literature. Extending Stilgoe et al., Scherer and Voegtlin articulate a "responsible governance" framework that incorporates the PCSR approach pioneered by Andreas Scherer and Guido Palazzo based on Jürgen Habermas's theory of deliberative democracy Scherer et al., 2016).
Whereas the highly influential Friedman doctrine calls for a strict division of labor between government and business, Scherer and Palazzo (2011) argue that in today's "postnational constellation" companies do and should "engage in self-regulation to fill global gaps in legal regulation and moral orientation" (p. 899). Since these actions are clearly political, companies in this competing view become political actors-they assume political corporate social responsibilities (Scherer & Palazzo, 2007). While the resulting research program is heterogenous, a dominant sub-stream focuses on the role and practice of corporate norm-making (de los Reyes et al., 2017). The core idea in this stream is that companies should live up to their political responsibilities and create norms to regulate business where there are governance gaps; favored examples in this literature include the multi-stakeholder deliberation that created the certification standards of the Forest Stewardship Council (Scherer & Palazzo, 2007, p. 1110) and the Accord on Fire and Building Safety that responded to the Rana Plaza disaster in Bangladesh (de los Reyes et al., 2017).
In discussing the governance forms of responsible innovation, Voegtlin and Scherer have emphasized open-systems approaches to governance that embed the corporation in a soft law network structure whose sum is superior to, and apparently autonomous from, its private sector parts (Voegtlin & Scherer, 2017). Leadership, as in the example of the Californian water management institution Calfed (pp. 239-240), is orchestrated and facilitated by governments and non-governmental organizations through alliance structures, and Voegtlin and Scherer envision the corporation within this web as an active agent observing and participating in public discourses with accountability to stakeholders (e.g., 2017, pp. 239-240). More specifically, PCSR scholars understand that to sustain their legitimacy, corporations from time to time have to embark on soft law norm-making initiatives that involve the corporation in democratic and discursive mechanisms of will formation that frequently span borders (Scherer & Palazzo, 2007, p. 1098Scholz et al., 2019).
The issues triggered by big tech innovation, like causing widespread addiction (Abroms, 2019;Bhargava & Velasquez, 2020), often do transcend borders, but the most significant regulatory voids also occur within the jurisdiction of governments with powerful and effective regulatory structures. The regulatory voids often faced by tech companies are not necessarily a function of an institutional void, such as arguably surrounded the safety of the Bangladeshi textile industry: "instead of encountering institutional voids (Khanna & Palepu, 2000;Uzunca et al., 2018), platformbased sharing economy business models encounter thickets of overlapping regulations that span multiple levels and jurisdictions and are not necessarily tailored to the sharing economy" (Garud et al., 2020). This is not to say applicable regulations are developed in a competent or timely way. In any event, in our focal case of Uber, the global regulatory standing of its business model is extraordinarily complex and localized: the Wikipedia (2002) entry on the 'Legality of ridesharing companies by jurisdiction' currently has over 8000 words, covering over 40 nations and dozens and dozens of political subdivisions. In New York City, the site of the CPA tactics we analyze, its mayor and city council posed a serious threat to Uber's growth strategy in 2015 with a growth cap that, if enacted, the company could not have evaded. Regulatory law can be at its strongest in a case like that, which is a significant difference with the poorly enforced building inspection requirements that should have prevented the Rana Plaza building in Bangladesh from ever being erected on the soft ground that caused its collapse.
Consider that in facing the European Union's comparatively demanding standards on issues like political advertisements on social media platforms (Collins, 2020), Facebook, Google, and Twitter find themselves pushed by hard law to observe more demanding standards than in the United States, triggering these companies to engage defensively and proactively to preserve the liberty to do business as they wish. That tactical, political space where the companies that host innovative technologies struggle to survive and thrive in hard law regulatory environments is the subject of this article. Looking at the focal case of Uber and how it counterattacked the possibility of a New York City growth cap, our question is not what CPA scholars might seek to answer, e.g., what makes a corporation good or not good at surviving the political environment, but rather we are concerned with the question raised by PCSR scholars: what would it take to pursue these ends in a morally legitimate way? Or more precisely: What makes CPA tactics morally (il)legitimate, why, and under what conditions? Our inquiry and analysis is applied and limited to the case of liberal democracies with political institutions and legal regimes that are relatively stable and enjoy reasonably wide legitimacy, as with the United States and countries in the European Union, leaving to future research the study of our research question in the important context of fragile states (e.g., Banerjee et al., 2021). Similar questions were raised by communication scholars  who propose a conception of "deliberative lobbying" that puts into practice PCSR's "theory of deliberative democracy, [whose] normative demands of political CSR can be summarized as discourse, transparency, and accountability" (p. 416; cf. Anastasiadis et al., 2018, p. 216). Lock and Seele argue that "existing lobbying strategies should be applied under the normative premises of political CSR (i.e., discourse, transparency, and accountability), which, in Habermasian terms, implies a shift from strategic to communicative action" (p. 9). Though they do not operationalize the requirements of discourse, transparency, and accountability, Lock and Seele suggest that, for example, lobbying tactics like "astroturfing" do not qualify see Anastasiadis et al., 2018;Schultz & Seele, 2020).

Letting Corporate Activities Speak for Themselves
Before presenting the focal case of Uber in New York City and answering the research question, this section will address a puzzle raised by the CPA and the PCSR literatures. Specifically, both conversations (CPA and PSCR) could be read to suppose that moral legitimacy is not the point of CPA tactics (e.g., Hillman & Hitt, 1999;Hillman et al., 2004;Patzer et al., 2018;Scherer, et al., 2013), which raises the doubt whether our research question is quixotic. To clearly situate our contribution as an extension of these literatures, we should respond to the concern that our research question is confused about CPAs as a phenomenon.
With respect to the CPA literature's supposition that companies engage with the political sphere only "to shape government policy in ways favorable to the firm" (Hillman et al., 2004, p. 838), we reject that position and instead embrace PCSR's view that companies can, may, and should act as political actors (Scherer et al., 2014;cf. Cashore et al., 2021). We take a different tack to respond to the PCSR literature's apparently similar assumption that CPA tactics like "advertising campaigns... lobbying, and other instruments of strategic public relations" pursue a "strategic manipulation strategy" that features "the dissemination of (mis)information" (Scherer, et al., 2013, p. 264). We suggest that PCSR scholars' bright line categorization of corporate political engagement into distinct types-CPA tactics that go no further than corporate interests and forms of deliberative engagement that do (Scherer, 2017, p. 388)-represents a simplifying assumption that is useful for theory building (e.g., Scherer et al., 2013, p. 264;see Aspalter, 2020). This duality runs deep into political CSR's foundations in Habermas (1987), who differentiates the deliberative "communicative action" that transpires in what he calls the "lifeworld" from the instrumentalist "strategic action" that plays out in the "system" (Patzer et al., 2018, pp. 329-330). PCSR theorizes about corporate activities through the prism of this duality: the deliberative democracy instituted through, e.g., the Forest Stewardship Council, is viewed as communicative action worthy of PCSR, whereas CPAs correspond to the instrumentalism of strategic action (e.g., p. 330).
As useful as a sharp distinction between strategic action or manipulation and the deliberative democracy of communicative action has been for theory development, we argue that differentiating corporate activities in terms of corporate motivation and intent, i.e., why companies engage in political practices to influence public policy, is difficult if not impossible to do. The problems are not only methodological. It is notoriously difficult to test a person's, let alone an organization's, true intention or motivation (see generally Orts & Smith, 2017). Moreover, as organizational research shows in abundance, organizations are not motivated by one reason at a time (e.g., profit maximization), but by bundles of sometimes even contradicting reasons and commitments (Besharov & Smith, 2014;Smith & Besharov, 2019; see generally Bower, 1970).
In any case, the phenomenon that both the CPA and PSCR literatures actually theorize about is not corporate intention but rather about corporate activities-the variety of tactics that corporations engage politically, from public relations campaigns to multi-stakeholder initiatives. We follow suit, and by bracketing motivations and theorizing about the moral legitimacy of CPA tactics as such, we believe we open a fertile channel for the PCSR and CPA literatures to cross-fertilize concepts and empirical insights. 2

Uber in New York City
Our theoretical inquiry is motivated by a phenomenon that has been widely recognized by the general public as well as scholarly commentators: highly innovative companies, especially so-called big tech, push the frontiers of technology and in so doing frequently enter markets in ways that clash with existing regulation; they do not ask for permission to enter first, and if anything they ask for forgiveness after the fact (Pollman & Barry, 2016, p. 398;Shahani, 2014). Having situated our project in the business ethics and management literature, we now present a concise review of the 2015 episode that pit Uber against the cap on rideshare vehicles championed by New York City Mayor Bill de Blasio. Uber's aggressive and defensive maneuvers in 2015 are striking for the dramatic and favorable reversal of political fortunes achieved that year with a full complement of CPA tactics that featured a highly innovative and effective grassroots campaign that tapped into the attention span and consumer interests of voters using the Uber rideshare app.
Uber launched its service in New York City in May 2011. Rides per day grew at a linear rate from below 100 k rides daily in 2015 to over 500 k rides daily in 2019 (Schneider, 2016). By 2015, Uber was growing by 30 k new users per week (Griswold, 2015). Rideshare vehicles (Uber and Lyft) became ubiquitous on the streets of NYC with over 60,000 cars, more than four times the cap on licensed yellow taxis (Rogers, 2015). In response to the surge of Uber's seemingly unfettered growth, Mayor Bill de Blasio spoke out against congestion and began a campaign to cap the number of rideshare vehicles on the roads, proposing a 1% per year limit on growth of Uber vehicles in New York City (Smith, 2015).
Uber reacted aggressively to Mayor de Blasio's campaign, pursuing CPAs on three major fronts: (1) indirect lobbying of the public via a variety of advertising channels and media outlets, (2) direct lobbying of city councilors (Sethi, 1982), and (3) constituency building of its own customers to achieve a powerful form of voter-driven lobbying of the city council. A central player in Uber's political strategizing recounts the arsenal of CPA tactics deployed: (1) generate massive public opposition to the bill through TV ads, radio ads, banner ads, rallies, newspaper support, pundit support, clergy support, community support, driver support, and support from elected officials (the outside game), and (2) conduct an intense lobbying campaign to somehow line up 26 'No' votes in the [city] council, constituting a constant barrage of calls, emails, and tweets from constituents, direct mail in each district either praising the councilmember for opposing the bill or attacking him or her for supporting it, polling to show the bill was unpopular, and having our team of lobbyists suffocate each member and their staffs (the inside game) (Tusk, 2018).
Uber's CPA strategy was expensive and "waged by lobbyists and strategists trained in the regimes of Obama, Cuomo, and Bloomberg" (Smith, 2015). Uber communicated the message that its "product has some genuinely progressive effects," and its ad blitz "highlighted how Uber's drivers are mostly black and brown [and] show[ed] that Uber serves outer-borough neighborhoods that for years were shunned by yellow cabs" (Smith, 2015). Uber enacted this campaign at a time when "Mayor Bill de Blasio believed that the pieces needed to score a quick win were in place" (Tusk, 2018).
In addition to traditional forms of indirect and direct lobbying, Uber mobilized its riders to lobby directly to city councilors through the innovative deployment of its smartphone app to reach the captive audience of its customers waiting for a ride (Fig. 1). Through this tool, Uber mobilized city residents to speak out as voters in a manner that blurs the line between indirect and direct lobbying. Riders using the Uber app were prompted to click on the "de Blasio" alternative to UberX and UberPOOL to learn "what Uber will look like in NYC if Mayor de Blasio's Uber cap bill passes" (Griswold, 2015). Users clicking de Blasio mode ( Fig. 1) would see much increased wait times with an invitation to "Email the Mayor and City Council. Say 'NO' to de Blasio's Uber!" The New York city council was bombarded with over 20,000 emails in five days (Pollman & Barry, 2016, p. 388). The pressure worked, and it was voiced by customers who spoke out as political citizens.
Uber's customer constituency building went beyond the virtual de Blasio mode campaign. The firm also decided to "offer[] free rides to passengers willing to attend a protest at City Hall on its behalf" (Pollman & Barry, 2016, p. 388). No doubt, the de Blasio mode's automated letter campaign garnered gravitas through live protest.
Uber also reached the broader public through indirect lobbying via the city's newspapers. "The editorial pages of all three city dailies labeled de Blasio's cap plan a bad idea" (Smith, 2015). Uber's NY general manager published an open letter to Mayor de Blasio that challenged the mayor in an opinion piece in one of the daily newspapers for having multiplied the rationales, beyond congestion, for wanting to cap rideshare vehicles (Bhuiyan, 2015). The letter requested further public discourse by "inviting him to a public, livestreamed conversation to discuss his proposal to cap the number of drivers of Ubers and other for-hire cars in New York City" (Bhuiyan, 2015). As Uber ramped up the pressure "with more TV, more radio, more calls, more lobbying, more public pressure," city hall sought a meeting, but when the bill proposing the cap was not dropped, Uber said it would only meet in a live stream so people could know what the contest was about. Uber and the city's government reached a détente: "We just promised to end the campaigntake down the TV ads, no more radio ads, no more mail, no more email or tweets" (Tusk, 2018). Uber agreed to a fourmonth traffic study "to examine the impact of Uber and the for-hire vehicle industry on traffic congestion on New York City streets" (Bhuiyan, 2015;Griswold, 2015). 3 In summary, Uber's successful counterattack responded to the threat of regulation with three key sets of CPA tactics (Table 1) to (re-)shape the regulatory regime in NYC to allow the market for its innovative business model to continue to grow.

Assessing the Legitimacy of CPA Tactics
Having described a specific case of a big tech company using CPA tactics to influence its regulatory environment, we have in view the kinds of phenomena that the analytical framework we now develop must make sense of. Our strategy is to start from the foundational yet indefinite concept of input legitimacy in political CSR and sharpen it by adding more detail to the theory of how stakeholders can be included in political discourse (Dimension 1: Stakeholder inclusion) and to the modes of communication and decision-making that have the capacity to channel discourse (Dimension 2: Modes of communication and decision-making). We will argue that to assess the legitimacy of CPAs the tactic(s) at hand should be assessed in relation to the scoring on these two dimensions.

Dimension 1: Stakeholder inclusion
The PCSR literature has stressed that the participation of corporations in public will formation discourses is sufficient for moral legitimacy if and only if these engagements satisfy certain rules of discourse: "the legitimacy of a political decision rests on the discursive quality of the decision-making process" (Scherer & Palazzo, 2007, p. 1107. Habermas (2005, p. 89) who is the spiritus rector and provides the philosophical foundation of PCSR, defines four such conditions for a legitimate discourse: (i) no one capable of making a relevant contribution must be excluded, (ii) participants have equal voice, (iii) they should be free to speak their honest opinion without deception or self-deception, and (iv) there is no coercion built into the process and procedures of the discourse. These conditions articulate a conception of what it means to assess all the relevant information and arguments as reasonably as possible, allowing for the force of arguments to rest on their merits in a disinterested and public pursuit of truth (Bohman & Rehg, 2014). 4 In an effort to operationalize Habermas's work for assessing the legitimacy status of private governance through multi-stakeholder-driven soft law, business ethicists Mena and Palazzo (2012) refer to these criteria of a legitimate discourse as the "input legitimacy conditions," where "input legitimacy" refers to a norm's "rule credibility" "by virtue of its processual history-who talked to whom, when, where -and how" (Scholz et al., 2019). Within the normative dimension of input legitimacy, the (1) "who talked to whom" and the (2) "how" factor are of particular relevance. As for the (1) who, stakeholder representation is of the essence (Reinecke & Donaghey, 2021). While Habermas (2006) and subsequent generations of discourse ethicists understood that the goal of perfect representation is idealistic and arguably unrealistic, the normative goal of aiming for high levels of representation remains intact. This is why Scherer and Palazzo (2007, pp. 1104-1105 establish PCSR according to Habermas's more recent account of deliberative democracy rather than the earlier conception of ideal discourse (Scholz et al., 2019, p. 323), and they explicitly demand that "those being affected by a norm must be able to participate in a real argumentation regarding its validity" (Gilbert & Behnam, 2009, p. 216). Similarly, in their contributions to responsible innovation, Scherer andVoegtlin (2018, 2020;cf. Voegtlin et al., 2018;Voegtlin & Scherer, 2017) embrace the view that the legitimacy of these processes is dependent in large measure on who participates in what ways to govern the innovation at hand.
We agree with this tradition that the inclusion of stakeholders in political processes is necessary to enhance the moral legitimacy of rules (Mena & Palazzo, 2012, see, e.g., Habermas, 1998Risse, 2004;Scharpf, 1999;Young, 2004). We hold that the legitimacy of CPA tactics that aim to influence public policy depends on the degree to which those affected by them are included to participate (see also Young, 2004). In other words, we suggest that the legitimacy of CPA tactics that aim to influence public policy can be assessed by looking at how well affected stakeholders are included (Dimension 1: Stakeholder Inclusion) and also to what degree a company is willing to engage in discourse with these stakeholders to decide on further action (Dimension 2: Modes of Communication and Decision-Making). Our theoretical contribution is to propose how to operationalize a principle for systematically assessing the legitimacy of CPA tactics more definitely than existing approaches and as applied to the political space that CPAs operate in and seek to shape. We borrow from and apply the rich multidimensional framework that political scientist Archon Fung (2006) developed to assess the relative legitimacy and effectiveness of the "governance sphere" (Cashore et al., 2021) of public participation of citizens in various forms of direct democracy. Fung (2006) is concerned with institutional design: how can the mechanisms of public participation in democratic formation will be designed to address the problems of "illegitimacy, injustice, and ineffectiveness of particular clusters of governance arrangements" (p. 66)?
The gap public participation is poised to fill concerns the unavoidable epistemological constraints of public officials (in our treatment, management). Whereas Fung articulates dimensions along which public participation in government action can be enhanced, we are able to extend his framework to CPAs because, as we demonstrate, companies face the same parameters for designing CPA tactics and, therefore, have to make choices along a spectrum of inclusion from no stakeholder inclusion to selective recruitment or even inclusive recruiting. Figure 2 depicts these degrees along an axis. 5 CPA tactics in general can be scored on a stakeholder inclusion scale, as we will demonstrate with application to the case of Uber in New York City. We argue that CPA tactics that are conducted with no stakeholder participation do not meet a minimal threshold for legitimacy, whereas CPA tactics that recruit stakeholders, whether selectively or inclusively, provide potential foundations for the legitimacy of CPA tactics that can meet the standards of responsible governance and innovation Voegtlin et al., 2018;Voegtlin & Scherer, 2017) (see Fig. 2).
No Stakeholder Inclusion. The least inclusive way for a company to design a CPA tactic that is meant to influence public policy and thereby shape regulation is to leave stakeholders-"any group or individual who can affect or is affected by the achievements of the organization's objectives" (Freeman, 1984, p. 46)-totally outside the process enacted by the CPA to shape public policy. Uber's indirect lobbying efforts, using television, radio, and banner ads, sent a message designed to influence what the city council would do, based on a theory of the impact of the messaging on its targets. Uber in no way instituted a process designed to provide Uber any stakeholder feedback to help Uber define in a participatory way what it was trying to achieve through the CPA (e.g., Uber, 2018).
We stress that we see an important difference between Uber, or any other company, targeting and addressing stakeholders. While the former primarily aims to use stakeholders as a means to a company's end (as marketing tactics do), the latter at least bears some potential for dialogue between the stakeholders and the company at hand. The difference between targeting and addressing stakeholders was recently at issue in Philip Morris International's campaign to run "sponsored content" articles in leading newspapers that appear as bona fide journalism but instead carries the company's message. For example, a sponsored content article in The Boston Globe blamed "misinformation" for barring the company from selling its products (Dimitri & Stinson, 2021). Philip Morris did not open a discourse by ensuring its sponsored content was juxtaposed with a response from public health advocates. Just like Uber's New York City messaging campaign, Philip Morris's message was entirely monological, designed to affect public policy outcomes by shaping public opinion.
Selective Recruiting. A more inclusive option to design a CPA tactic is what Fung calls Selective Recruiting. This mode is different from the above for two reasons: First, corporate officials aim not only to target but also to address a sub-set of stakeholders. Second, the corporate officials live up to the normative demand of including stakeholders by recognizing the potential impact of innovation on these groups of individuals. The company at hand then addresses a selected sub-set of these stakeholders directly. Building on instrumental stakeholder theory, a company could choose to only address stakeholders that have powerful, legitimate, and urgent claims and thus count as what Mitchell et al., (1997, Fig. 2 Dimension 1: stakeholder inclusion 5 To be sure, the extent of a legitimacy and governance gap that corporations may need to bridge through public participation is a function of the company's corporate governance structures and processes. For example, German companies arguably benefit from co-determination to better discern good answers to the managerial puzzle underlying our research questions, namely, how to legitimately-and effectively-reshape the regulatory environment to accommodate the company's business model and strategy. Nevertheless, co-determination's capacity to strengthen the legitimacy and effectiveness of a company's political strategies depends upon the quality of not only the structure of governance but also the processes, including how well the respective power of workers' and shareholders' representatives, respectively, are integrated into decision making. Thus, both Fung's and our frameworks ask not only who participates (Dimension 1) but how (Dimension 2). p. 874) term definitive stakeholders, while ignoring dependent stakeholders that may also have urgent and legitimate claims but lack power.
When Uber used direct lobbying to influence how city councilors would vote on Mayor de Blasio's desired rideshare vehicle cap, the company used selective recruiting and created a channel with city councilors, who are important stakeholders with a legitimacy enhanced by their status as elected representatives. To take another case, hospitals in the United States that are organized through the American Hospital Association (AHA) recently lobbied the federal agency that administers the Medicare program (national insurance for the elderly) "to repeal the requirement that hospitals and health systems disclose privately negotiated contract terms with payers on the Medicare cost report" (Nickels, 2021). The AHA did recruit the federal agency; however, the AHA's selectivity did not involve the elderly in particular or taxpayers in general, though implicated by the norm-making strategy.
Inclusive Recruiting. The most inclusive way to fulfill the normative demand not to exclude anybody who can make a relevant contribution (Habermas, 2005;Mena & Palazzo, 2012) is to design CPA tactics that aim for "descriptive representativeness" (Fung, 2006, p. 68). In this mode the CPA tactic addresses not only the definitive (or some other sub-set of) stakeholders but all stakeholders affected by the legal regulation (or its absence) implicated by the CPA tactic in question. While it might be illusory to include all affected groups or individuals, companies that operate in this mode seek descriptive representation of stakeholders. All who wish to engage with the topic at hand are invited to do so.
A company can offer all stakeholders an open virtual forum to comment (or even to decide) on its future course of action, as seen in the case of Ben & Jerry's grassroots lobbying for the People's Response Act. The company provided information on the purpose of this proposed legislation and asked all web page visitors to call their representatives in Congress (Ben & Jerry's, 2021).
We stress that complete openness to input from any stakeholder who steps forward is, on its face, an attractive vehicle for democratic participation, and yet its vulnerability in practice is to the "self-selection" of stakeholders who actually shape the CPA tactic. Writing about open citizen forums in the context of public policy, Fung (2006) warns that "those who choose to participate are frequently quite unrepresentative of any larger public. Individuals who are wealthier and better educated tend to participate more than those who lack these advantages, as do those who have special interests or stronger views" (2006, p. 67;cf. Fiorina, 1999). Fung's reservation concerning self-selection mechanisms can be applied to CPA tactics as well. It is unrealistic to think that all affected stakeholders could equally participate in whatever kind of dialogical process a company offers. Moreover, Reinecke and Donaghey (2021) warn against "self-appointed representatives, including campaign groups and non-governmental organizations (NGOs), [who] participate in transnational rule making on behalf of citizens, consumers, local communities, workers and others in largely unregulated global supply chains but lack an explicit mandate to represent" (p. 1).
To address these difficulties and to make participation more attractive to those who are ordinarily less likely to participate, structural incentives may be provided (for example by selective online and offline communication, selection of venues, etc.). Uber, for example, shuttled its users for free if they were willing to protest in person. Taking a different tack, a company could institute an inclusive and selective process by randomly selecting stakeholders among the general population. This practice is already well known in public policy initiatives such as deliberative polling, Citizens Juries, and Planning Cells that randomly select participants to discuss public issues (Fishkin 1995;Gastil 2000;Fung, 2006;Leib 2010;Smith & Wales 2000).
Whether companies employ selective recruiting or inclusive recruiting, the moral legitimacy advantage these approaches have in practice next to non-inclusive CPA tactics will depend on tactical design and execution. Notwithstanding these complications, and as indicated above, we suggest that the moral legitimacy of a CPA tactic can be evaluated along the dimension of stakeholder inclusion. The more inclusive a CPA tactic is the better the claim that it is legitimate (Table 2).

Dimension 2: Modes of Communication and Decision-Making in CPA tactics
While we hold that stakeholder inclusion is a necessary dimension to assess the legitimacy of a CPA tactic, it is not sufficient to do so-that is why any conclusions based on Dimension 1 address only the potential for legitimacy. To assess the legitimacy of a CPA tactic, one also needs to ask how the company interacts with its stakeholders via the tactic at hand.
In a perfect world, approximating a discursive ideal, the stakeholders would reason with each other directly as equals without (bargaining) power imbalances and would give preference to the best arguments (Habermas, 1984(Habermas, , 1987. While these were the suggestions of the early Habermas, the later Habermas (1996) admitted that these conditions were too idealistic. Habermas in his later works emphasizes the concept of deliberative democracy (Habermas, 1996). It is this concept of deliberative democracy that has been dominantly picked up by PCSR scholars (Gilbert & Behnam, 2009;Habermas, 1996;Scherer & Palazzo, 2007;Scholz et al., 2019). Scholars from this domain frequently argue that the deliberative engagement of corporations with stakeholders in multi-stakeholder initiatives (MSIs) serves to legitimate corporations in their role as political actors (Scherer & Palazzo, 2007;cf. Scholz et al., 2019). They choose to build from deliberative democracy (i.e., the later Habermas) not because they doubt the capacity of ideal discourse to ground moral legitimacy (Hussain & Moriarty, 2018), but rather because ideal discourse "provid[es] a more utopian than realistic orientation for corporate behavior" (Scherer & Palazzo, 2007, p. 1105. What mainly distinguishes the early and the later Habermas on this view is the willingness under the latter to relax the strictures of the ideal speech situation. Seen thus, discourse might be possible even if stakeholders can hardly ever deliberate truly as equals, without any (bargaining) power imbalances. Consensus is not absolutely essential so long as the discursive process yields a rational basis for disagreeing about the moral remainder (Hursthouse, 1995). Mena and Palazzo (2012) began to operationalize these conditions for a legitimate discursive process in arguing that power differences in decision-making structures should be neutralized (procedural fairness); stakeholders should uphold a culture of cooperation and reasonable disagreement (consensual orientation); and structures, processes, and results should be transparent (transparency). Notwithstanding these normative background assumptions, business ethics scholars so far have not developed a scale to assess the different modes of communication and decision-making in any domain of norm-making, much less for CPA tactics.
We argue that the modes of communication and decisionmaking within CPA tactics can be distinguished based on the quality of the dialogue and the degree of stakeholder participation in corporate decision-making processes. Applying Fung's (2006) typology, we see the potential for CPAs to range from the very passive mode of constituency engagement Fung calls "Listen as Spectator," to the more engaged mode he labels "Express and Develop Preferences," and to the significantly more intensive modes of "Negotiation," and at the top of the scale, proper "Deliberation." Fig. 3 depicts these modes on the scale from least intensive to most intensive in terms of dialogical participation and decision-making with stakeholders.
Often CPA tactics are purely monological-the company seeks to put out a message for political influence through one means or another. Whoever receives the message being broadcast is there to listen as a spectator, and the company expects the spectator to be moved in some expectedly advantageous way. In this space, the targets of the corporation do not gain a channel to reciprocally engage in any kind of dialogue with management; instead, they receive information about some policy or project, and sometimes they bear witness to struggles among other involved parties (e.g., corporate representatives, politicians, activists, and interest groups).
Uber's indirect lobbying campaign through TV, radio, and banner ads went no further than putting stakeholders in the position to listen as spectators to the messaging designed to create political pressure to vote down the vehicle cap in the city council. Another example can be found in a CPA tactic of the tobacco industry. Doctors from Tobacco Free Massachusetts criticized The Boston Globe for running paid content from Philip Morris arguing, first of all, that the newspaper's 1999 ban on tobacco advertisements should have barred the content. In addition, the criticism was that Philip Morris's speech went unanswered in the newspaper: A Philip Morris executive is unchallenged when she says that her company is using "science" to solve the problem of cigarette smoking. There is no counterpoint showing that her company caused and continues to promote the very problem she says the firm is solving. There is no information challenging the effectiveness or safety of the new product she promotes (Dimitri & Stinson, 2021).
Had Philip Morris or the Globe required a counterpoint to take up stakeholders' perspectives, the CPA tactic could have gone further than Listen as Spectator.  Sometimes, CPA tactics provide opportunities for stakeholders to express their preferences to the company in question and further to develop or to change these preferences, be that in a live setting, for example, by an open microphone, or via a comment function in a virtual arena (see Schultz & Seele, 2020). Here companies encourage participants to learn about issues and, if appropriate, transform their views and opinions by providing them with background educational materials or briefings and then asking them to consider the merits and trade-off s of several alternatives. The CPA tactic opens space for stakeholders on and offline to discuss these issues among other stakeholders and with the company. In this mode, stakeholders are not merely the targets of an information deposit (as in the listen as spectator mode).
Examples range from corporate social media channels with a comment function (e.g., Twitter, Facebook, YouTube, etc.) to private governance initiatives in which non-industry stakeholders are invited to voice their concerns (without necessarily being included in decision-making). FedEx took to Twitter in 2017 to call for tax reform as Congress was about to pass the Trump tax cuts (Public Affairs Council, 2018). Another example is the Canadian Chemical Producers Association (CCPA) which, following the world's worst industrial accident at the Union Carbide plant in Bhopal India, created the "Responsible Care Initiative" (RCI) and "assembled a cross-sectional group of activists, academics, consumers, seniors, and youth who followed the development of Responsible Care from its inception" (Donaldson & Schoemaker, 2013, p. 31). In this way, the RCI invited stakeholders to express preferences that informed its policy making.
Note that CPA tactics that deploy stakeholders to listen as spectators and even to express and develop preferences create at best highly diffuse engagement. As Fung explains, these modes of communication are not designed to "translate the views or preferences of participants into a collective view or decision" (Fung, 2006, p. 68). In the resulting fora, government officials or, as in the case of CPAs, companies "commit to no more than receiving the testimony of stakeholders and considering their views in their own subsequent deliberations" (p. 68).
Moving higher on the scale, we see CPA tactics that do attempt to develop a collective choice. In the mode of negotiation, the company engages with stakeholders in a space where the stakeholders in question have preferences and bargaining power that demand negotiation to arrive at a collective choice. The exploration and give-and-take of bargaining can allow participants to find the best available alternative to advance their respective joint preferences (Fung, 2006, p. 68).
For example, behind the deadlock among World Trade Organization members over the waiver of patent protection for COVID vaccines under the Agreement on Trade-Related Aspects of Intellectual Property Rights, there was a negotiation between pharmaceutical companies and the United States government, which over the summer of 2021 shifted from being likely to approve the waiver (Lawder, 2021) to deciding against it (Farge, 2021). Another iconic example is seen in classical lobbying where a company directly and openly approaches government officials to gain subventions in return for, e.g., building a factory and creating jobs in the jurisdiction. Here two parties (i.e., the company and the government) engage in an open bargaining game and the outcome reflects the differential bargaining power of the two parties (Helpman & Persson, 1998).
Negotiation represents one way for both the company and its stakeholders to engage in dialogue and to participate in shaping the outcome of a CPA tactic. However, the highest quality public participation as agreed by Fung (2006) and the Habermasian tradition in business ethics (PCSR) involves inclusive stakeholder participation and deliberation. In this mode, participants not only negotiate but deliberate with each other to determine what they want individually or as a group (Fung, 2006;Scherer et al., 2013). In governance forms that enable deliberation, stakeholders take in educational background materials, exchange perspectives and experiences, and reason with one another to develop their views and discover their interests. In the course of developing their individual views in a group context, deliberative mechanisms often employ procedures to facilitate the emergence of principled agreement, the clarification of persisting disagreements, and the discovery of new options that better advance what participants value . The deliberative mode can be distinguished by two features. First, a process of interaction, exchange, and-ideally-edification precedes any group choice. Second, participants in deliberation aim toward agreement with one another (though frequently they do not reach consensus) based on reasons, arguments, and principles (see Habermas, 1996;Scherer & Palazzo, 2007).
For example, the Accord for Fire and Building Safety in Bangladesh is widely considered a prime case of highly inclusive stakeholder representation as well deliberation and thus a case of deliberative democracy. After the Rana Plaza building collapsed, representatives of global labor -Bangladeshi unions along with buyer companies from Europe, North America, and Asia-reached a collective agreement through negotiation and deliberation resulting in the formation of this initiative. Additionally, labor rights NGOs were involved as witness signatories (Reinecke & Donaghey, 2021).
As indicated above, we suggest that the moral legitimacy of a CPA tactic can be evaluated along the dimension of communication and decision-making. Other things equal, the more dialogical potential a CPA tactic enjoys, the more power imbalances between participants are resolved, and the more any decision approaches deliberation, the more legitimate is the CPA tactic at hand (Table 3).

Applying the CPA Legitimacy Framework
In summary, we argue that the legitimacy of a CPA tactic for responsible innovation can be assessed along two dimensions (Dimension 1: Stakeholder Inclusion; Dimension 2: Modes of Communication and Decision).
We suggest that to assess the legitimacy of a given CPA tactic one should rate the tactic along the two of dimensions of (1) Stakeholder Inclusion and (2) Modes of Communication and Decision. These dimensions are depicted in the following framework (Fig. 4). We suggest that other things equal the higher a CPA tactic scores on each dimension, the higher its moral legitimacy status (Fig. 5).

Evaluating Uber's CPA Tactics in New York
As we saw above, CPA and PCSR scholars alike have understood that companies engage in CPA tactics primarily to promote corporate interests (Hillman et al., 2004;Scherer et al., 2013). For PCSR scholars, this tarnishes and undercuts the tactic's moral legitimacy status (Scherer et al., 2013, pp. 263-264;Rasche, 2015;Scherer, 2017, p. 388;Scherer et al., 2006Scherer et al., , p. 511, 2009Scherer et al., , p. 330, 2014Scherer & Palazzo, 2008, p. 421, 2011Scherer et al., 2016, p. 274). We recognize that highly innovative companies employ CPA tactics for a host of diverse reasons, including profit maximization. Notwithstanding their profit orientation, some management teams might also be driven by a logic of "self-defense" when engaging in CPA. If companies face inflexible and outdated public policies, responsible innovation calls for companies to find legitimate ways to influence these potentially problematic public policies. In what follows, we will apply the CPA legitimacy framework to assess three main sets of Uber's CPA tactics in NYC (Table 1).

Traditional Indirect Lobbying Through TV Ads, Radio Ads, Banner Ads
One set of Uber's CPA tactics was designed to generate massive public opposition to Mayor de Blasio's proposed rideshare cap using TV ads, radio ads, and banner ads. To apply the CPA legitimacy framework, we look to the dimension of (1) stakeholder inclusion as well as (2) modes of communication and decision-making (Uber 2018) for a representative television advertisement). Ostensibly, this tactic is fully inclusive and yet scores low because the effort to influence the political process does not include stakeholder input at all-stakeholders are not addressed in the CPA tactic (Dimension 1) but rather are targeted. There is no dialogue to assess on Dimension 2; these tactics are purely monological without any listening from the side of Uber: the citizen listens as involuntary spectator. Accordingly, this set of tactics fits in the lower right corner of the CPA legitimacy framework (Fig. 5). The tactics that belong in this quadrant do not meet minimal standards for potential legitimacy in shaping law. 6

Classical Lobbying of City Councilors
Another heavily used CPA tactic in the 2015 Uber episode was classical lobbying. Classical lobbying aims to affect public policy by providing or exchanging information formulated by an interest group (i.e., a company) (Lock & Seele, 2016, p. 213). Hillman and Hitt (1999) characterize lobbying as an "information strategy" in the toolkit of CPA tactics because it aims "to affect public policy by providing policy makers-specific information about preferences for policy or policy positions and may involve providing information on the costs and benefits of different issue outcomes" (p. 834). With regard to this tactic, we suggest that its legitimacy strongly depends on how it is exercised. If the company openly approaches city hall members (i.e., policy makers) in order to inform these stakeholders on the company's contribution (e.g., providing a truly innovative ride service that strongly enhances convenience for customers, bring people into jobs, is non-discriminatory, etc.) and its needs (i.e., regulations that allow the innovation provided by Uber), and if it engages in negotiation or even deliberations, this tactic enjoys comparatively high legitimacy (Drutman, 2015;cf. Lock & Seele, 2016 for their differentiation of instrumental and deliberative lobbying). From the perspective of communication and decision-making (Dimension 2), the company informs the stakeholder-the policy maker who acts as an elected representative of other stakeholders-and helps them to develop and to express preferences and then engages in a form of negotiation or even deliberation. From the perspective of stakeholder inclusion at least one stakeholder-the politician-is addressed. Since the city hall members in turn enjoy legitimacy because they have been elected via a democratic process and thus represent multiple constituents, it might be argued that this tactic scores even higher on Dimension 1. However, that higher ranking for stakeholder inclusion assumes that the form of argumentation appeals to democratic interests. If the lobbying is a negotiation with the legislator with respect to the legislator's personal interests (rather than the polity's), this kind of lobbying (for which there is no evidence in this case) might fit in the upper left zone of the CPA legitimacy framework (bargaining without genuine stakeholder inclusion). Transparent lobbying based on arguments about the policies and their impacts for stakeholders more broadly is the most legitimate version of lobbying. 7

Constituency Building of Customers / Grassroots Campaigns
Arguably, the most original of Uber's 2015 tactics is a form of grassroots lobbying. In this instrumentally effective form of lobbying (Lord, 2000;Nownes, 2006), "lobbyists do not approach the public policy makers directly, but take a detour via the constituents of a certain community or issue field to reach the politician indirectly (Lock & Seele, 2016, p. 3;cf. Thomson & John, 2007). Companies try to "influence public policy by gaining support of individual voters and citizens, who, in turn, express their policy preferences to political decision makers" (Hillman & Hitt, 1999, p. 834). Uber did exactly this. By triggering de Blasio Mode and app-generated customer letters to city council, Uber's tactic can be seen as indirect lobbying via constituency building of its customers through a (monological) grassroots campaign. Here Uber engaged directly with customers and indirectly with the city council to inform these stakeholders on the consumer implications of the mayor's public policy plan to marshal support for Uber's desired outcome. With this tactic, Uber directly and indirectly addressed multiple stakeholders, customers, and city council. While the facts do not suggest that Uber mobilized stakeholders more generally, the tactic as deployed scores relatively well on the dimension of stakeholder inclusion. However, the way that Uber mobilized its customers was not designed to spur feedback or critical engagement from them-no dialogical potential is indicated. While it can be argued that Uber educated its stakeholders about potential policy changes and their implications for Uber and its customers, Uber's engagement through this tactic remained purely monological. We therefore suggest that this set of tactics belong in the bottom right quadrant of the legitimacy framework.
The following figure summarizes our assessment by applying the legitimacy framework to the three sets of Uber's CPA tactics that we examine in particular (Table 1).

Discussion and Conclusion
In this article, we focused on the phenomenon of responsible innovation. While the seminal literature theorizes the factors that render innovation responsible  or how responsible innovation can be governed Voegtlin & Scherer, 2017), we look at the phase after a company has launched its innovative product or service. After this initial phase, the company is often confronted with strong negative reactions from its stakeholders. While some stakeholders might positively embrace the innovation, others might be more skeptical and even start to oppose it. This phase of backlash is dangerous for the innovating company. If protests are growing and the public policy maker reacts with regulation that cuts the business opportunity down, the company might lose the potential for competitive advantage that its innovation created. While we do not question the primacy of elected governments to set the rules of the game for business, we align with those scholars who argue that with highly innovative or even disruptive services and products, policy makers are sometimes overwhelmed and not competent enough to set policies that protect the public interest without "killing" the innovation at hand (Van Waarden, 2001;cf. Andrews, 1972, p. 143). We believe that companies have the right to defend themselves against the sometimes incompetent, inflexible, or unbalanced regulation that may issue. In fact, we wish to bolster the view that companies are legitimate stakeholders in public discussions with a right or even a responsibility to participate in processes that can support or shut down their innovations (see also Locke & Seele, 2016, p. 9).
Not surprisingly, companies already do actively engage themselves in the political arena (Pollman & Barry, 2016). When they experience a backlash from stakeholders, and especially by public policy makers after entering the market to launch their innovation, companies start to counterattack. For these counterattacks, companies have several tactics at their disposal as we have shown above. The aim of this paper has been to develop and operationalize a legitimacy framework with which to assess tools companies can legitimately use to protect their business interests in the political sphere. Considering Uber's aggressive response to the political backlash against the vertiginous growth of its innovative service, which of the CPA tactics it employed are legitimate? Scholars have begun to examine the relationship of PCSR and CPA in different ways (Anastasiadis, 2014;Anastasiadis et al., 2018;Den Hond et al., 2014;Rhodes & Fleming, 2020). Our article contributes to this nascent research stream with an innovative theoretical framework, and we show how the two streams may be aligned to the benefit of both.
As indicated above, Scherer andPalazzo (2007, 2011) build upon the literature on corporate citizenship (Matten & Crane, 2005) to make the case for the genuine political responsibility of companies. Going further than business ethics scholars who portrayed companies as norm-takers (e.g., Donaldson & Dunfee, 1999;cf. de los Reyes et al., 2017), Scherer and Palazzo defined a far more politically active role for business. Against the backdrop of the long list of pressing social and environmental problems implicated by business practice, they suggest that companies should become political actors and engage in MSIs to set new or better rules of the game.
What distinguishes us from Scherer and Palazzo's PCSR, and from Voegtlin and Scherer with respect to responsible innovation, is that we propose a clear assessment framework to evaluate the legitimacy of the political activity of companies deploy to defend their innovations. Thus, our article contributes to the literature on the debate on the relationship of CPA and PSCR in the context of innovation. , 2017 raised the potential for companies to engage in responsible lobbying, suggesting lobbying can be seen as legitimate when it satisfies the conditions of discourse, transparency, and accountability (Lock & Seele, 2016, p. 416). We go beyond their work by proposing a more fine-grained and integrated framework for the legitimacy assessment of CPA tactics.
Moreover, we suggest that our framework can be applied to more than the assessment of CPA tactics used by companies in the course of innovation. As we demonstrated with the discussion of the Bangladeshi Accord, our framework applies to assess the legitimacy of the MSIs that Scherer and Palazzo propose can address the global governance gap (for examples, see Mena & Palazzo, 2012), and the forms of governance that Voegtlin and Scherer characterize for responsible innovation because we use the same criteria-input legitimacy-to assess the legitimacy of these institutions. While these authors, especially Mena and Palazzo (2012), started to operationalize input legitimacy based on the factors of inclusion, procedural fairness, consensual orientation, and transparency (p. 527), we have extended beyond their efforts by providing a nuanced framework for the assessment of MSIs (see endnote 1).
While the foundation of our framework-PSCR-has been extremely influential in business ethics and other disciplines, it is not without critiques. One of the most stingingand pertinent-critiques was recently advanced by Rhodes and Fleming (2020; see also Dawkins, 2015Dawkins, , 2021aDawkins, , 2021b. In their article "Forget political corporate social responsibility," the authors argue that PCSR has been "hailed by many as a solution to societal problems not dealt with by government" (2020, p. 943), but that it actually "reflects both a triumph of neoliberal corporate power and a harbinger of democracy's demise" (ibid.). Their worry is that PSCR is not only a smoke screen for companies to continue with business-as-usual but that it is a predatory corporate project. They fear that "rather than governments functioning as a public counterbalance to unbridled capitalism, [PCSR] permit[s] firms to govern themselves because the state is too ineffectual" (Rhodes & Fleming, 2020, pp. 945-946). The realm of responsible innovation is not spared by this critique-neither Voegtlin and Scherer's proposal for how to govern these innovations nor our analysis of the postinnovation phase that leads companies to unleash an arsenal of CPA tactics to defend the market for their innovative business models. Both Scherer and Voegtlin as well as the authors of this article suggest that companies should actively engage in norm-making processes to govern innovation whether through voluntary initiatives Voegtlin & Scherer, 2017;Voegtlin et al., 2018) or by actively co-creating local and national hard law regulation (the authors of this article). We share Voegtlin and Scherer's premise that companies must do so, not only because of the post-national constellation and the corresponding global governance gap, but primarily because governments sometimes lack the technical knowledge and the capacity to quickly set adequate rules of the game (cf. Andrews, 1972;Bower et al., 2011).
Our assessment framework sheds light on Rhodes and Fleming's (2020;cf. Branicki et al., 2021;Goodman & Mäkinen, 2022) criticism that companies that interfere with politics are dangerous, especially when they are big and powerful tech companies. To use a metaphor, if governments allow these companies to use all their power, i.e., all the CPA tactics that are available to influence public policy, it is like letting a 500 pound gorilla into a crowded supermarket. The gorilla will not ask nicely for a banana but will just grab whatever it wants. Uber's actions in NYC are analogous here. This big, powerful tech company used its metaphorical muscles (i.e., CPA tactics) to successfully influence public policy. Can we endorse the way Uber did this? Our answer is no, and while that view may be widely held pre-theoretically, our analysis provides systematic reasoning to discern the reason why. The assessment framework assumes that the domain of politics cannot be entirely off limits to CPA tactics but should allow reasonable self-defense. There are responsible and legitimate as well as illegitimate ways to wage those campaigns, and we have provided an assessment tool for differentiating CPA tactics accordingly.
However, as we stated above, companies are arguably needed to help public policy makers set the rules that allow for both corporate success and innovation on the one hand and social and environmental sustainability on the other. Rather than condemning the political activity of companies altogether, as Rhodes and Fleming seem to do, we suggest that policy makers should find ways to drive companies to employ CPA tactics that score high on our legitimacy assessment framework. At the same time, the policy maker might think of legally banning or disincentivizing tools that score low on the assessment framework, since these powerful tools can undermine democracy. In discussing whether companies should or should not act politically, we suggest demurring to the general question and instead looking into the concrete CPA tactics at issue. We hope to have provided useful criteria of which tactics fail to meet minimal standards and which might be considered legitimate.
In addition, we suggest that companies should take a closer look at the legitimacy status of the CPA tactics they use. After the open feud between Uber and New York City Mayor Bill de Blasio, the de Blasio administration agreed to drop its proposed plan to impose a cap on the rideshare service's annual growth. The leading business press outlets reported, "Big win for Uber, New York backs down" (O'Brien & Segall, 2015;Tracy, 2015). It seems as if Uber's 2015 use of its CPA tactics in NYC were indeed successful. However, just three years later the de Blasio administration struck back. "New York became the first major American city […] to halt new vehicle licenses for ride-hail services, dealing a significant setback to Uber in its largest market in the United States" (Fitzsimmons, 2018). In 2021, the divide between the company and policy makers seems to be bigger than ever. Regulators around the globe, partially spurred by Uber's aggressive use of CPA tactics, regulated Uber so heavily that in some cities, Uber has become almost indistinguishable from a traditional taxi service. While many factors, including the aggressive behavior of Uber's former CEO Travis Kalanick, leads to the impression of Uber as the epitome of big, bad tech (Naughton, 2014;see McCardle, 2019), the company has arguably brought constraining regulations upon itself by aggressively deploying CPA tactics that enjoy a shaky legitimacy status. We, therefore, propose that companies consider using our framework to proactively analyze the legitimacy status of their CPA tactics. The alternative is to risk widening the antagonism between the company and policy makers, threatening their innovation and thus their competitive advantage.

Declarations
Conflict of interest All Authors declare that they have no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visithttp:// creat iveco mmons. org/ licen ses/ by/4. 0/.