Civil society in times of change: shrinking, changing and expanding spaces and the need for new regulatory approaches

The relationship between many G20 governments and organized civil society has become more complex, laden with tensions, and such that both have to find more optimal modes of engagement. In some instances, state-civil society relations have worsened, leading some experts and activists to speak of a “shrinking space” for civil society. How widespread is this phenomenon? Are these more isolated occurrences or indeed part of a more general development? How can countries achieve and maintain an enabling environment for civil society? The authors suggest that much of the current impasse results foremost from outdated and increasingly ill-suited regulatory frameworks that fail to accommodate a much more diverse and expanded set of civil society organizations (CSO). In response, they propose a differentiated model for a regulatory framework based on functional roles. Based on quantitative profiling and expert surveys, moreover, the paper also derives initial recommendations on how governments and civil society could find ways to relate to each other in both national and multilateral contexts. (Published as Global Solutions Paper) JEL F5 L31 H7 K33

• Finally, there is the policy perspective that views nonprofits as a source of social innovation in addressing diverse public problems. 1 Indeed, nonprofits are assumed to be better at such innovations than governments typically are: their smaller scale and greater proximity to communities affected and to those concerned makes them creative agents in finding solutions. They are the operating ground for social entrepreneurs. Governments are encouraged to seek a new form of partnership with CSOs aimed at identifying, vetting and scaling up social innovations to build more flexible, less entrenched, public responses.
Importantly, these perspectives cast CSOs in strikingly different roles. At one level, they become parallel actors that may substitute, even counteract, state activities. At another, the state and CSOs are part of ever more complex and elaborate public-private partnerships and typically work in complementary fashion with other agencies, public and private. Civil society harbors significant potential in terms of social innovations, resilience, service-delivery and giving voice to diverse interests and communities otherwise excluded. However, CSOs operating locally, national and across borders have also experienced many changes in recent decades. The current decade has brought about a particularly complex and challenging environment (Anheier 2017). Specifically, there are growing indications that the "space" for civil society organizations is shrinking worldwide as a result of increased regulation, greater reporting requirements, but also curtailing of CSO activities, and even harassment of staff and threats of violence among growing authoritarianism (Carothers and Brechenmacher 2014;Civicus 2018;ICNL 2018;USAID 2017).

Approach, Data and Findings
To assess the state of civil society across the G20 countries, and, particular, to probe how widespread the shrinking of civil society space has become, we use data available from the international social sciences project Varieties of Democracy (V-Dem) (see https://www.vdem.net/en/). Specifically, we chart the space for civil society organizations over time along three dimensions (Coppedge et al. 2018): • Control over the formation of civil society, which measures government discretion in granting legal status to CSOs (CSO Entry and Exit); • Control over the operations of civil society, which measures bureaucratic harassment and repression of existing CSOs (CSO Repression); and • Degree of Self-organization and Participation, which measures CSO diversity and voluntary participation (CSO Participatory Environment) We also differentiate the G20 member countries by political regime type, using the Economist Intelligence Unit's (EIU) Democracy Index: _________________________ www.economics-ejournal.org 4 • Full Democracies (Australia, Canada, Germany, United Kingdom); • Flawed Democracies (Argentina, Brazil, France, India, Indonesia, Italy, Japan, S. Korea, Mexico, South Africa, United States); • Hybrid Regimes (Turkey); and • Authoritarian Regimes (China, Russia, Saudi Arabia).
The purpose here is to show how the space of civil society has changed in the course of the last decade, i.e., from the global financial crisis of 2008 to 2016. In a second step, we look into the policy context to gauge how countries manage to balance the potential civil society offers with the mandate of governments of state and international organizations to serve as keepers of peace and arbiters between major political and economic interests. Among the results, as presented below, several stand out: • The V-Dem data do indeed suggest a general, but mostly gradual erosion of civil society space: values measuring freedom from government control over the entry or formation or exit or dissolution of CSOs are, on balance, lower in 2016 than they were in 2008 ( Figure  1). The same holds for government repression and self-organization and participation as well (Figures 2 and 3). While these values are lower, they are not lower in the sense that they would have dropped suddenly or by much. Nonetheless, the overall trend suggests some gradual erosion rather than dramatic decline. The main exception is Turkey, the one hybrid regime in the G20, where the situation for CSOs rapidly deteriorated after 2010 (B panels in Figures 1-3). • The few G20 countries that show overall improvements along the three indicators are the four full democracies in the EIU index (D panels in Figures 1-3). The great majority, however, reveals a pattern of either relative stability or gradual decline across the dimension of civil society space. While this is not surprising for the authoritarian regimes (A panels), which were at low levels already, more worrisome is that civil society space contracted in several flawed democracies to significant degrees in recent years, which contributed most to the overall slow erosion. • This suggests that some democracies may at least not actively seek to develop civil society space through reform efforts. Instead, they more or less passively let civil society space slowly erode either through the impact of other policies (mostly anti-terrorist, anticorruption, and national security related legislations and measures) or lack of reform. 2 It also suggests that hybrid and authoritarian regimes are the clearest case of a shrinking (e.g., Turkey) and shrunk (e.g., Saudi Arabia, Russia, China) civil society space, whereas for flawed democracies, it would be better to speak of a slow process of erosion. _________________________ 2 A prominent case in point are Financial Transactions Task Force efforts to curb international money-laundering and terrorism financing that have effectively hindered or even cut off access by NGOs to banking and other financial services and are considered a key contributor to shrinking spaces for CSOs in many advanced democracies (Human Security Collective & European Center for Not-for-Profit Law 2018). The extent of the financial access problems for charitable nonprofits in the US was empirically documented in a 2016 Schar School survey (Daigle et al. 2016). Beyond this, Sidel (2006) discusses the impact of security and anti-terrorism policies in the aftermath of 9/11, and the more recent securitization of Chinese NGO policies (Sidel 2018).
Economics: The Open-Access, Open-Assessment E-Journal 13 (2019-8) Global Solutions Papers www.economics-ejournal.org 5 To probe deeper into these issues, we asked a group of civil society experts (see Appendix 2) three questions: • What are the main challenges for CSOs, both domestically and in terms of cross-border activities, and what opportunities present themselves? • What are likely trajectories for CSOs over the next five to ten years, especially with changing geo-politics? • From a policy perspective, what could be the roles of national governments and international organizations in that regard? Are reforms and models of state-civil society relations being discussed?
We also asked if, in the course of the past five years or currently, changes to, or new, laws and regulations have been put in place or are being passed or envisioned that either facilitate and improve or complicate and worsen the establishment and operations of: • domestic CSOs; • international CSOs headquartered abroad and working in the country; • domestic CSOs working internationally.
Appendix 1 presents a synopsis of answers received along three dimensions: the state of civil society, the implications for its expansions, stability or contraction, and the need for reform and dialogue. While Appendix 1 offers a rich portrait of the diversity of civil society, its relationships with governments, and its trajectories across G20 countries, there are also four overarching results: • the general trajectory of a slow erosion in most consolidated democracies is confirmed, as are the developments in hybrid and authoritarian regimes, although the expert reviews add important nuances; • few countries have open, proactive dialogues in place to review civil society -government relations; the most common pattern is the absence of a policy engagement rather than some form of contestation; • fewer countries still have reform efforts under way, even though a general sense of reform needs prevails among expert opinions; • most countries seem to do little to stem the erosion, perhaps out of unawareness, lack of civil society activism and organizational capacity to find a common voice, or the absence of political will on the part of governments.
More specific results are: • There are characteristic "pendulum policies" in a number of G20 countries with more pronounced differences between center-right and center-left governments that tend to politicize the relationship with civil society and contribute to inconsistencies over time; • Several G20 countries have seen the need to respond to the hybridization of CSO, especially around service-provision, and are establishing new forms like social enterprises or public benefit corporations as part of an effort to modernize regulatory frameworks; • Government bureaucracy is seen as a major stumbling block to more efficient relations, especially in middle-income countries; there is a need to simplify registration processes and reporting requirements in particular; in some countries, registration is also used as a tool to control CSOs and restrict their activities; • Few countries have umbrella organizations for CSOs, which leads to disjointed civil society voices, and decreases advocacy capacity; • Some countries establish dedicated government agencies for CSO oversight, control, and also development.
Thus, CSOs find themselves in contradictory policy environments across the G20 member countries, and subject to a 'push and pull' along the different directions by the challenges and opportunities the perspectives above harbor. In some cases, the contradictory policy environment stems from inattention and a lack of understanding how other policy measures such as anti-terrorist legislation affects civil society. In others, the expert assessments provided in Appendix 1 suggest that countries with shrinking or eroding spaces are also those in need of regulatory reforms. This is consistent with the general shrinking space experience, which is largely based on the promulgation of restrictive laws and regulations (Rutzen 2015).

A Need for New Regulatory Approaches
The limited and outdated policy approaches and regulatory frameworks in many countries (Phillips and Smith 2011) are a key impediment to improving the development trajectories for civil society. In essence, no G20 country has an explicit, normative approach concerning civil society to guide regulatory frameworks that could help realize CSO potentials. Instead, regulation is either almost exclusively fiscal in nature and rests on some notion of public utility that CSOs serve; or is controlling in the sense that state authorities oversee nearly all aspects of CSO operations and governance. The former, as typified by the US, implies some form of a 'light' hands-off regulatory framework with few general government supports other than tax benefits. The latter represents a stricter hands-on regime that encompasses the shrinking space problem, albeit with more financial and other contributions by the state for qualifying CSOs, that can simultaneously see their space expanding. Emerging exemplars of this approach are the dual government postures towards NGOs in Russia Benevolenski and Toepler 2017) and China (Zhang 2015). For the fiscal regulatory regime, the key governance question becomes: is the organization entitled to preferential tax treatment; and for the control regime, it is: does the organization fit into government policy and state-determined priorities? Clearly, most G20 countries fall somewhere in between but are closer to the fiscal framework. These regulatory regimes are of course embedded in larger policy rationales governing the relationships between government and CSOs.

Regulatory Approaches and Government-Nonprofit Relationship Models
Economic theory offers three answers to this question, each casting CSOs in a different role (see Steinberg 2006;Anheier 2014: Chapter 8, 16). Young (2000) in particular has suggested a triangular model of government -civil society relations of complementarity, substitution, and adversity. He argues that to varying degrees all three types of relations are present at any one time, but that some assume more importance during some periods than in others. It is the task of policy to balance this triangle. The notion that CSOs are supplements and substitutes to government rests on the public goods and government failure argument first advanced by Weisbrod (1988): they offer a solution to public goods provision in fields where preferences are heterogeneous, allowing government to concentrate on median voter demand. CSOs step in to compensate for governmental undersupply. The theory that CSOs are complements to government was proposed by Salamon (1995), and finds its expression in the third-party government thesis whereby CSOs act as agents in implementing and delivering on public policy. Indeed, we find that even hybrid and authoritarian regimes, such as Russia, have developed a strong interest in involving CSOs in service-delivery with state support (e.g., Benevolenski and Toepler 2017). CSO weaknesses correspond to strengths of government (public sector revenue to guarantee nonprofit funding and regulatory frameworks to ensure equity; and CSO strengths (being closer to actual needs, more responsive) complement government weaknesses. A basic laissez-faire fiscal regime corresponds well with these models, although it is often enriched with additional incentives (eg, grants, contracts) to encourage nonprofit participation in cooperative relationships with government. Expanding spaces for civil society occur where supplementary or substitutional models are in place.
More controlling regulatory regimes, by contrast, are typically motivated by conflictuous relations. The theory that CSOs and governments are adversaries is supported by public goods arguments (see Boris and Steuerle 2006) and social movement theory (Della Porta and Felicetti 2017): if demand is heterogeneous, minority views may not be well reflected in public policy; hence self-organization of minority preferences will rise against majoritarian government. Moreover, organized minorities are more effective in pressing government (social movements, demonstration projects, think tanks) than unorganized protests; however, if CSOs advocate minority positions, the government may in turn try to defend the majority perspective, leading to potential political conflict. To an extent, a similar dynamic also underlies the current global backlash against foreign funding of mostly social justice and rights-focused advocacy NGOs that has been at the heart of the shrinking space phenomenon (Christensen and Weinstein 2013;Dupuy et al. 2016).
Associating regulatory regimes with these relationship models points to some, though not all, of the deficiencies of prevailing policy approaches that limit the developmental potential of CSOs. Specifically, we see three key deficiencies: First, CSOs have different organizational forms and governance structures: • The membership association, based on some shared interests of members that as demos form the basis for its internal governance; • The non-profit corporation based on set capital and limited liability, where a board substitutes for owners and represents their interests; and • The foundation, an ownerless asset dedicated to a set purpose, and a board functioning as trustee.
Social entrepreneurs cut across these forms and provide the 'active ingredient' for innovations and development (Brewer 2016;Young et al. 2016). With some exceptions, regulatory regimes generally fail to take these differences in form and governance into account.
Second, CSOs, and the entrepreneurs and employees as well as members and volunteers operating in them, perform different functions or roles that allow them to realize their comparative advantages (Kramer 1981): • Service-provider role: substituting or complementing services offered by government and businesses, often catering to minority demands, and providing trust goods (high information asymmetries and high transaction costs), thereby achieving an overall more optional level of supply; • Vanguard role: less beholden than business to the expectations of owners demanding return on investment, not subject to shorter-term political success, and closer to the front lines of many social problems and needs, CSOs can take risks and experiment, thereby increasing the problem-solving capacity of society as a whole; • Value-Guardian role: fostering and helping express diverse values (religious, ideological, cultural etc.) across a population and within particular groups when governments are either constrained by majority will or autocratically set preferences, thereby contributing to expressive diversity and easing of potential tensions; • Advocacy role: when governments fail to serve all needs and groups in the population equally well, and when prevailing interests and social structures can disadvantage certain groups while given unjust preference to others, CSOs can serve as public critics and become advocates, thereby giving voice to grievances, reducing conflicts and possibly effecting policy change.
These special roles for nonprofits are also not fully reflected in regulatory regimes. Considering the relationship models though, the service provision and vanguard/innovation roles associate closely with the supplementary and substitutional models and the opening and expansion of spaces for civil society. The value guardian and advocacy roles, by contrast, are reflective of the adversarial model, controlling regulation and shrinking spaces for civic activity.
While NGOs can bring advantages, the third deficiency is the failure to recognize and address the inherent weaknesses of CSOs, including (Anheier 2014): • Resource inadequacy, whereby the goodwill and voluntary contributions cannot generate resources adequate and reliable enough to cope with many of the problems facing member states. • Free-rider problems, whereby those who benefit have little or no incentive to contribute, stand in the way of sustainable resourcing, too. • Particularism, whereby CSOs focus on particular subgroups only while ignoring others, which can lead to service gaps; conversely, if CSOs serve broader segments of the population, they encounter legitimacy problems. • Paternalism, whereby CSO services represent neither a right nor an entitlement but are at the discretion of particular interests that may not necessarily reflect wider social needs or the popular will. • Accountability problems, whereby CSO, while acting as accountability enforcers and pushing transparency, are themselves inflected by such insufficiencies.
The resulting challenge is clear: how can the advantages CSOs bring be strengthened while minimizing any disadvantages, and while taking differences in form into account? What is the right policy framework and regulatory approach to balance the respective interests of governments and civil society while realizing the potential of civil society? Current frameworks seem unable to achieve such a balance without further differentiations.

Needed Differentiation in Policy Frameworks
The main proposal for finding proper policy responses to these issues is that a more differentiated approach to CSOs is needed, and one that goes beyond the one-size-fits-all of current regulatory frameworks. These are largely based on some notion of charity and public utility, and have a regulatory history reaching back to the late 19 th and early 20 th century, and in some cases even to mediaeval times. They are rooted in outdated notions of how organisations should serve the public good, and they fail to consider the diversity of modern organizational forms and ways of collective action. Instead, frameworks should be based on the functional differentiation embodied in the policy models above, taking into account organizational forms, comparative advantages and drawbacks.

CSOs as service providers
The first differentiation is for CSOs in the service providing role. A proper regulatory framework should differentiate entirely charitable, donative CSOs from CSOs that are part of public-private partnerships, from those participating in quasi market arrangements with competitive bidding for fee-for-serve contracts, and, more generally, from CSOs that operate in competitive fields alongside public agencies and businesses. In the large and growing fields of education, health and social care, CSOs face many fiscal problems and limitation in making business decisions in keeping with their nonprofit status, while businesses accuse them of unfair competition due to tax exemption. Most CSOs here are corporations given the significant capital requirements rather than membership-based associations, but they have virtually no access to capital markets for investments and cannot compete for talent against businesses able to offer more competitive compensation packages. As a consequence, many CSOs push against regulatory boundaries that may threaten their tax status (Weisbrod 1998;Toepler 2004b, Eikenberry andKluver 2004).
The main regulatory issue is to establish workable ways of oversight in relation to the forprofit -nonprofit borderline, and hence to facilitate access to capital markets. New hybrid legal forms are currently being devised to solve some of the underlying issues, suggesting the need for a more differentiated system. The L3C and the benefit corporation in the US or the public benefit corporation in the UK are steps to fix various shortcomings of both the nonprofit and for-profit forms (Brewer 2016;Abramson and Billings, in preparation). Reflecting this need for differentiation, some observers have proposed the concept of a fourth sector comprised of "forbenefit enterprises" (Sabeti 2011).

CSOs as an expression of civic engagement
A second differentiation addresses the function of CSOs performing the advocacy and value guardian roles, typically in the form of associations. Here the main regulatory issue is between primarily self or member-serving activities, on the one hand, and ensuring accountability on the other. Democratic legitimacy frequently gets called into question when representation issues arise. Many of the democratic legitimacy issues being raised about both local and international CSOs have to do with membership and community representation (Brechenmacher and Carothers 2018). In addition, even in the West, there is a troublesome decline in active association membership, as members frequently chose not to participate in the 'schools of democracy' aspects of democratic decision-making, including internal elections and attendance at membership meetings.
CSOs seeking to advance specific member interests frequently confront charges of putting their particular benefit above others and see their beneficial tax treatment questioned and their motives challenged. This has been a particularly salient issue for economic associations, such as cooperatives and mutual societies (Salamon and Sokolowski 2016). What is needed is a regulatory framework that recognizes different degrees of publicness versus privateness of the interest pursued: primarily public-serving objectives should be treated in a beneficial way, while member-serving ones may not. Many interests will fall in between, and these should only receive partial benefits. Importantly, financing of political parties should not be regarded as part of civil society and regulated separately, including the activities of political action committees and similar vehicles that channel private funds to the world of politics.
Beyond the problems resulting from interspersing party politics and charitable nonprofits, the regulation of political activities, such as advocacy and lobbying, is another major area of regulatory concern, especially in Anglo-Saxon countries. In the US, potentially draconian tax penalties for possible violations of vague lobbying rules have for decades hindered the willingness of charities to even engage in legitimate advocacy activities despite clear evidence of its usefulness. High-performing nonprofits utilize service-providing expertise to leverage their advocacy and employ advocacy to improve services and the general policy environment for their clients and constituents (Crutchfield and Grant 2007). Here it is both the political activities and the party politics versus civil society border that needs better regulation.
Civic engagement also places CSOs, across all forms (and increasingly also Internet-based advocacy platforms), in the role of social accountability enforcers (Fox 2015, Brinkerhoff andWetterberg 2016); again making them vulnerable to charges of lacking transparency and catering to special interests. Needed here are a higher degree of accountability standards, including transparency for themselves (Ebrahim and Weisband 2007;Gugerty and Prakash 2010). What is more, the profound changes in conventional media and the cacophony of social media resulted in a loss of standards and professionalism and brought with them a weakening of the public sphere in many countries, as well as a loss of trust in institutions. Here, regulation is needed that establishes minimum public transparency and accountability requirements while aiming at improving the quality of the public sphere, to the extent that workable systems of selfregulation cannot be encouraged (AbouAssi and Bies, 2018).

CSOs as financial intermediaries
A third differentiation is about private support for the public good, which foregrounds the roles and potential contributions of philanthropic foundations. Foundations endowed with incomegenerating assets are generally considered to be among the most unconstrained institutions in society, as they are neither beholden to market expectations nor to the electoral booth. This dual independence from economic and political considerations allows them to address complex, controversial, even unpopular issues, and seek solutions where government and business are likely to falter, let alone risk taking them on in the first instance. Foundations can take the longer view and operate without regards to shorter term expectations of market returns or political support. Accordingly, foundations are primed to pursue a set of special societal roles, including pursuing change and innovation, redistribute wealth, build out societal infrastructure and complement, or substitute for, government action (Anheier and Hammack 2010;Anheier and Leat 2018). Unfortunately, governments often fail to understand appropriate foundation roles and primarily look to them as mere 'cash machines' to fill emerging gaps in public budgets (Abramson et al. 2014;Toepler 2018) or tend to overregulate them (Leat 2016;Toepler 2004a). Prewitt (2006) has argued that foundations in liberal societies allow attaching private wealth to the pursuit of public goods with only limited interference in economic choice and political freedoms. Striking a balance between the two is a key regulatory challenge.

CSOs as social innovators
The fourth differentiation is about social innovation, and applies to corporations, associations and foundations alike. CSOs do function as innovators and vanguards yet they face fundamental problems in terms of replicability, diffusion and scaling up ). There is no systematic screening and vetting of social innovations, and many fail due to inadequate dissemination and information-sharing. As a result, the potentials of too many social innovations go unnoticed, and 'wheels are being reinvented,' so to speak. And even those innovations that do find resonance, do so in the absence of a social investment market. Many innovations in civil society can harbour significant profitability for investors and owners as well as significant potential for the wider public -but in what direction the potential of a particular innovation will be realized in terms of replicability and scalability -and for whom -is often uncertain. Unlike in the case of technological innovations, there is no pool of investors eagerly standing by to help grow social innovations. Impact bonds and related measures are one step in the right direction (Albertson et al. 2018), but more is needed. Therefore, a platform or clearinghouse to assess any such potentials is needed, and a regulatory frame that would help social innovations to be tested. The organizational form and legal status of a platform or agency can be varied but should aim at establishing a social investment market next to the investment and venture capital markets for businesses.

Conclusion
Civil society, challenged in many ways, yet harboring huge potential, finds itself at a crossroads in many G20 countries. CSOs have long outgrown their regulatory frameworks, and it befalls to policymakers to provide adequate environments. The policy challenge is clear: How can the goals, ways and means of governments, and civil society be better coordinated and reconciled? What is the right policy framework to balance their respective interests while realizing the potential of civil society and taking account of the functional differences among CSOs and the various organizational forms underlying them? What rules and regulations, measures and incentives would be required? How can profoundly adversarial relations be transformed into complementary or supplementary ones without endangering the fundamental independence of civil society? Clearly, whatever answers will be found depends on the wider political and regulatory regime of each country considering these questions. They may be different for liberal democracies than for regimes in transition or failing states, let alone autocracies. Therefore, before the questions above can be addressed in terms of more concrete policy proposals, longerterm and ultimately normative issues must be considered: • Does a country see value in social self-organization generally, and, particularly, in a relatively independent civil society ready to challenge and confront those in power if need be -or does it prefer top-down social order with an emphasis on control? How countries settle the issues involved here will have major impact on any regulatory framework for CSOs as associations and their civic engagement function. • Does it view civil society services primarily as responses to government and market failures in delivering quasi-public goods when demand is heterogeneous -or are CSOs mostly extension agents of and for government, considered a versatile tool of delivering services governments might seek to off-load? Again, the kind of regulatory framework for serviceproviding CSOs as nonprofit corporate bodies serving the public good will depend on how countries come out in addressing these issues. • Does a country see value in independent forms of wealth dedicated to a public purpose as defined by private actors, and thereby possibly shape policy agendas -or does it view such activities are meddling with the public will, however defined, presenting a hinderance to governmental policies? The regulatory framework for foundations will be contingent on the outcome to that question. • Does a country see a main locus for social innovation in CSOs and the social entrepreneurs that act within and through them -or are such innovations primarily the responsibility of government planning and the interventions of state institutions? Again, how policymakers think about the functional location of social innovations will significantly influence any future regulatory framework, including any potential social innovation and investment markets.
All of these issues require fundamental debate considering the longer-term trajectories of government -civil society relations. They are likely to lead to preferences for different, and hopefully also improved, regulatory frameworks. Perhaps one reason for the often contradictory policy environments for CSOs, and the policy neglect they have encountered and are encountering, is perhaps that these long-term first-order governance decisions have been left unattended, even avoided, for too long.