Introduction

For the past two decades, rights-focused civil society organizations have been under legislative assault. With increasing speed since the late 1990s, governments around the world have been enacting new legal measures designed to suppress civil society’s functions and organizational space to carry out advocacy and politically oriented work. Coined as a global backlash against, or crackdown on, civil society—and by others as “closing civic space” or “shrinking civic space”—these legal measures are designed to restrict the emergence, activities, connections, and funding sources of civil society groups. A large number of the world’s countries have adopted new, more restrictive regulations to constrain the work of human rights-focused NGOs that receive funding from Western liberal democratic governments and liberal foundations in particular. This new approach to the regulation of civil society has attracted increasing attention from scholars, practitioners, and policy-makers alike (c.f. Lamarche, 2019; Chaudhry, 2022; Fransen et al., 2021; Bromley et al., 2020; Smidt et al., 2020; Bakke et al., 2020; Glasius et al., 2020; Poppe & Wolf, 2017; Dupuy et al., 2015, 2016; Christensen & Weinstein, 2013). Recent examples include India’s 2010 Foreign Contribution Act and Israel’s 2016 NGO Transparency Law.

We now have a fairly good understanding of which countries are adopting these legal measures as well as why they are doing so (Bromley et al., 2020; Chaudhry, 2022; Dupuy et al., 2016). More evidence continues to emerge about how civil society organizations try to adapt to and survive these regulations, and how various actors such as foreign aid donors and international institutions respond to and push back against these laws (Dupuy et al., 2015; Fransen et al., 2021). However, we still have not answered a basic question about the systematic, cross-national global effects of these legal restrictions on numbers of international non-governmental organizations (INGOs) in particular, since it is these organizations (and their funders) that have been heavily targeted. In this research note, we aim to contribute to this knowledge gap by examining whether and how restrictive civil society legislation shapes INGO numbers, and in particular, which type of legislation is impacting which types of INGOs. To do this, we use newly updated quantitative data on NGO restrictions and INGO numbers during the post-Cold War period to test hypotheses about the effects of regulatory restrictions on INGO numbers. Our quantitative results show that legal restrictions on advocacy activities (but not on funding sources) reduce the number of transnational human rights organizations, but not total numbers of INGOs per country, even when disaggregating by type of restriction.

We proceed as follows. We first briefly review the larger literature on the determinants of the size of the INGO sector. We give an overview of the context of the global backlash against civil society and the key findings to date on this topic in the academic and policy literatures. We then discuss our theoretical expectations regarding the impact of restrictive NGO legislation on INGO numbers. We present the data and method underlying our study and discuss our results. We find that total INGO numbers are not affected by restrictions. However, transnational human rights organizations numbers do drop, in particular because of government restrictions on advocacy. Our results therefore lend further support to (a) the position in previous qualitative studies that regulatory restrictions affect NGO types unevenly and (b) the perspective that for overall INGO populations, the effect of restrictions does not appear to be population “shrinkage.” We conclude by outlining a research agenda on the consequences of the ongoing global backlash against civil society.

Determinants of INGO Sector Size

As scholars like Reimann (2006), Keck et al. (1998), and Davies (2014) note, NGOs have become important players in world politics. Explanations abound for the size and shape of transnational civil society (see Davies, 2014; Anheier et al., 2020 for excellent overviews). Material explanations for this growth include the availability of generous funding from Western states in particular, which have tended to view NGOs as alternatives to dysfunctional states in the Global South to achieve development outcomes (Reimann, 2006). INGO numbers decline in over-populated organizational fields that strain available resources, due to competition over scarce resources and existing organizations actively dissuading new organizations from starting up (Bush & Hadden, 2019). Normative explanations focus on the role of INGOs in spreading pro-democratic norms and values, with globalization encouraging the rise and spread of “rational voluntarism” as a legitimate organizational form (Boli & Thomas, 1997). Societal structure, institutions such as education and democracy, and socio-economic relationships also seem to shape INGO numbers, including the degree to which a state is integrated into the world polity and international economy (Bailer et al., 2012; Lee, 2010; Salamon et al., 2017).

Most of the explanations reviewed above look to global forces to explain the size of the INGO sector; however, INGOs, like any other type of civil society organization, operate within national borders and as such, their ability to operate at all in a given geographic location is a function of the national regulatory framework governing civil society. In recent years, states have weaponized this framework to restrict the operations of foreign-funded and foreign-sourced NGOs, a form of governance scholars have referred to as “rule by law” (Scheppele, 2018).

The Global Backlash Against Civil Society and Its Effects

In this research note, we are concerned with the impact of national regulations on numbers of formally registered international non-governmental organizations (INGOs). Defining civil society or nonprofit organizations is a definitional and conceptual minefield (Anheier et al., 2020). We define NGOs as “formal organizations that are not directly part of government or the for-profit sector, and which seek to provide services to marginalized groups, and/or advocate for social or policy change” (Dupuy et al., 2015). INGOs are non-governmental, not-for-profit organizations that are usually headquartered in one country and have affiliated offices or branches in one or more additional countries. As formal organizations, they must have official recognition and approval to operate by the state, making them uniquely vulnerable to regulatory crackdowns, particularly regulations that seek to limit foreign influences.

The now well-documented increase in the adoption of national regulations restricting the emergence, operations, funding sources, and exit of INGOs is part of a larger phenomenon of a state-initiated and sanctioned clampdown on formal civil society groups and funding sources aimed at promoting human rights and democracy (Dupuy et al., 2016). A recent useful and parsimonious example of a typology of restrictions adopted thus far is that of Christensen and Weinstein (2013), later updated by Glasius et al. (2020) (and largely followed by Chaudhry, 2022) (see Table 1). This typology distinguishes between restrictions that pose barriers to entry, funding, and engaging in advocacy work.

Table 1 Restrictive NGO measures (Christensen & Weinstein, 2013; Glasius  2020)
Full size table

Studies have documented the effects of legal restrictions on organizations as well as the responses of (mostly Western, democratic) states as well as foreign aid donors, international institutions, and NGOs to this phenomenon (c.f. Fransen et al., 2021; Hossain & Oosterom, 2021; Tysiachniouk et al., 2018). Few studies have been carried out on the impact on organizational numbers; one example is Dupuy et al. (2015), who found that Ethiopia’s 2009 Charities and Societies Proclamation resulted in the death of nearly all of the country’s independent human rights organizations, and that other advocacy organizations generally switched to service activities to survive. In India, enforcement of the Foreign Contributions Act of 2010 forced international advocacy-focused NGOs like Greenpeace to close down due to the law’s restrictions on NGOs’ receipt of foreign funding, and the Indian government has used the law to cancel the operating license of thousands of other organizations (Swart, 2020). More research is needed to systematically identify how national legal restrictions shape INGO population ecology, including what restrictions seem to have most impact on NGO populations and in particular, on targeted populations of organizations like INGOs.

We expect that an increase in the total numbers of restrictive NGO regulations will negatively influence the total number of INGOs. This expectation is first based on a generic assumption about regulatory effects on populations of organizations, which are likely to raise the cost of operations, thereby leading to a reduction in numbers as a result of some organizations not being able to make ends meet. States are gatekeepers for INGOs; because INGOs operate within the borders of nation states, they must seek approval to carry out their work from the regulatory authorities (Hannan & Freeman, 1977). States regulate NGO barriers to entry, shape how NGOs can operate and organize themselves, determine the specific issues that NGOs can work on and who they can hire, govern how NGOs can mobilize and use funding and other resources, and regulate how NGOs can cease operations (Bloodgood, 2010; Henderson, 2010). This applies to INGOs as well; while all NGOs are vulnerable to shifts in domestic regulation, INGOs are perhaps more vulnerable than national NGOs to regulatory shifts that constrain the ability to access and use foreign-sourced funds, and to set up operations and engage in political advocacy as a foreign-based entity. Regulations aimed at limiting foreign intervention in national politics will, by nature, impact foreign-based organizations like INGOs, making them especially vulnerable. In line with the “shrinking civic space” metaphor, restrictive regulations are symbolic of a context that will be more difficult for a civil society organization to operate in.

Yet, there are also grounds to hypothesize that NGO numbers in a country would not fall as a result of regulations. If we understand the evolution of INGO populations as indicative of civil society and civic space, this position then represents a “transformation of civic space” perspective, rather than a “shrinking civic space” one. Some studies claim that many NGOs may continue work after regulation installed, while adjusting their activities or their organizational form (Dupuy et al., 2015). Others claim that such regulations, and more generally repressive government activities, harm certain categories of NGOs (mostly those critical of government regimes), while allowing other kinds of (more government-friendly or depoliticized) NGOs to flourish (Van Wessel et al., 2019; cf. Roggeband & Kriszan, 2021). NGOs disbanding may then be compensated by new and different NGOs emerging, leading to no clear observable effect in size of NGO populations overall. We in the end aim to empirically adjudicate between these rival claims of visible change in NGO numbers through our study, but theoretically consider a reduction in numbers most plausible for INGOs.

We further expect some types of targeted restrictions to have greater effects than others on numbers of specific types of organization, in particular restrictions on receiving funding from abroad. This is because many organizations can opt to switch away from rights-focused work or reframe such work as service delivery. Access to foreign sources of funding may, in fact, prove to be a source of resiliency; organizations can survive if they are able to maintain funding flows while adapting their operations to comply with new legal restrictions. Funding, however, is less substitutable than are particular issue areas of work, especially within larger organizations. And given the high dependence of formal INGOs on external sources of funding, particularly Western foreign aid, we expect that restrictions on funding would have the most significant impact on (1) INGO numbers, since these organizations depend heavily on foreign-sourced funding, and (2) on transnational organizations focused on advocating for human rights, since the work of these organizations is often the direct target of restrictive NGOs regulations and also generally requires foreign funding to carry out. In the absence of data on the actual implementation of restrictive NGO regulations (see the discussion below on limitations), observing the impact on the number of organizations most obviously and directly impacted by a regulatory crackdown can provide insights into the way in which restrictions are reshaping civil societies in adopting states.

Studying INGO Populations

There are important methodological limitations and issues to consider in studying INGO populations, namely how to conceptualize and empirically measure organizational numbers. Conceptual and normative definitions of organizational forms and structures (such as the definition we provide at the outset of this article) are different from legal definitions of organizational forms, which vary across national jurisdictions (Martens, 2002).

In our analysis, we use data from the Yearbook of International Organizations on numbers of international organizations to measure civil society numbers per country, per year.Footnote 1 The United Nations has mandated the Union of International Associations to publish the Yearbook on annual basis, collecting information from organizations via email, postal mail, and fax; Yearbook staff validate the information that they receive. The Yearbook defines an NGO as “a legally constituted organization created by private persons or organizations without participation or representation of any government. The term is usually used to refer to organizations that are not conventional for-profit business. NGOs can be organized on a local, national or international level (INGO).” Specifically, our measure of INGOs includes the following categories recorded by the Yearbook: federations of international organizations; universal membership organizations; intercontinental membership organizations; regionally defined membership organizations; and organizations having a special form, such as foundations. The Yearbook also records information on dissolved and inactive organizations, as well as on national organizations (though the temporal coverage for national organizations is more limited than for international ones). The Yearbook remains the best data source on numbers of organizations, although it privileges formal organizations and has a moderate response rate (35%).

Bloodgood et al. (2023) suggest that future research should employ alternative cross-national measures of NGO numbers, including using social media information, NGO reports, and a new dataset on non-profits based on national data sources like tax filings. Other scholars have used GIS data on NGO projects to estimate NGO activity (see Galway et al., 2012), while open-source project-level data that contains information on NGO implementing partners is also available from sources such as AidData,Footnote 2 Development Portal,Footnote 3 and Humanitarian Data Exchange.Footnote 4

Data and Method

For our statistical analysis, we rely on an updated version of the Glasius et al. (2020) panel dataset, which records the number and type of restrictive legislative measures imposed on NGOs in 96 countries between 1992 and 2018. This dataset is an updated and slightly adapted version of the dataset created by Christensen and Weinstein (2013). We have updated the Glasius et al. (2020) dataset with additional information on restrictions adopted for two additional years (2017 and 2018) as well as with information about international NGO (INGO) numbers per country-year using data from the Yearbook of International Organizations. The number of INGOs per country year serves as our first dependent variable. Our second dependent variable is the number of organizations per country that claim to have human rights as their main goal or primary issue focus; this variable is recorded every other year between 1993 and 2017 and is taken from the Transnational Social Movement Organization Dataset, which we updated to the year 2017 (Smith et al., 2017).

Our main independent variable is the total number of restrictive NGO regulations adopted per country-year. We update the Glasius et al. (2020) dataset covering such regulations to the year 2018 and code 11 types of legal restrictions under 3 broader categories. For each year and for each category of restriction, the 96 countries covered in the dataset are recorded as having either no restriction (0) or as having the restriction in question in place (1). In addition to testing for the effect of the total number of restrictions (a summary count of the total number of restrictions in place in each country, each year), we also test for the effect on INGO numbers of each of the category of restrictions in the Glasius et al. (2020) typology. These categories are measured as index variables that combine the measures under each of the three categories shown in Table 1; states can and do adopt multiple restrictions simultaneously within categories.

We control for a number of variables that influence INGO numbers. The statistical appendix contains details about all the countries and variables contained in our dataset and models (including control variables), shows descriptive statistics and the results of all statistical models run, and provides some additional statistical checks. We control at the country level for regime type, level of economic development, population size, education levels in the population, whether a country is experiencing armed conflict, and the amount of overseas development aid (ODA) flowing to NGOs operating in a given country. Of these variables, higher amounts of ODA and education as well as larger population sizes have a statistically significant effect on increasing the number of rights-based organizations in particular.

Empirical Analysis

To systematically analyze the impact of restrictive NGO regulations on INGO numbers and numbers of transnational human rights organizations, we use the Glasius et al. (2020) dataset and employ a time series linear regression model with fixed effects and robust standard errors clustered at the country level. Following the approach of Glasius et al. (2020) as well as other quantitatively oriented scholars in this field, we lag all the independent and control variables by one year so as to ensure they occur prior to the dependent variable.

Interpretation of Results

We run three sets of models but show only the most important results in Table 2 (all models can be found in the statistical appendix). First, we test for the effect of the total number of NGO legal restrictions on total numbers of active INGOs per country-year (Model 1) and total numbers of transnational human rights organizations per country (Model 2). Second and third, we re-run a similar analysis as in Model 1 but test instead for the effects of funding (Model 3) and then advocacy (Model 4) restrictions, again on total numbers of INGOs and human rights organizations. Only in Model 4 do we find a negative and statistically significant effect of advocacy restrictions on human rights organizations, not all INGOs.

Table 2 Multivariate regression results
Full size table

Contrary to our expectations, we do not find any significant, negative impact of the total number of restrictions on total numbers of INGOs, nor on specific types of NGOs like human rights organizations. Funding restrictions also do not have an impact on organizational numbers, whether all INGOs or rights-focused ones. Targeted restrictions that specifically limit the very bread and butter of international civil society groups—advocacy work—appear to make it harder for these groups to work in countries with such restrictions. Such groups are not likely to be able to switch to service delivery activities and maintain their advocacy focus and identity; in other words, human rights and service are not substitutable activities for this category of organizations. This finding is in part consistent with scholarship on NGO survival and the “transformation of civic space” perspective. This space does not necessarily shrink for INGOs when states adopt restrictive NGO legislation but rather changes the NGO (and also the INGO) landscape, allowing non-political (I)NGOs and regime-friendly domestic NGOs to dominate (Dupuy et al., 2015; Ljubownikow & Crotty, 2013).

Qualitative studies on the impact of NGO legal restrictions on rights-based organizations in particular illustrate how government clampdowns result in a reduction in these types of organizations, as they cannot easily substitute other issue areas. Organizations may choose to shut down in the aftermath of the adoption of new restrictions in order to skirt government attention, or they may be forced to close their doors by the authorities. As mentioned earlier, Dupuy et al. (2015) show how this occurred in the case of Ethiopia. Restrictive NGO regulations in Bangladesh also led some human rights organizations to actively cease operations to avoid government scrutiny (Fransen et al., 2021). And in another and more recent example, in 2021, the Belarusian government unilaterally dissolved dozens of human rights organizations, charging them with carrying out activities beyond their charters (Human Rights Watch, 2023).

Limitations of the Data and Analysis

Our results indicate no clear effect of regulations on overall INGO populations, lending support to perspectives in the literature that while change may occur as a result of restrictions, it does not necessarily mean a drop in numbers of organizations or activities. Our findings also indicate that regulatory restrictions do restrict the number of rights-focused groups. As such, our study confirms what country case studies have described previously for a larger group of country and NGO cases, but now across a range of different government regime types and stages of economic development. Next to this, our study, as an advance to the literature, points specifically to restriction of advocacy activity as the element of regulation that most strongly affect rights-focused NGO populations, as the large-N focus allows us to systematically compare different kinds of restrictions.

However, our analysis faces a few limitations that should be addressed in future research. Our independent variable measures adopted legal measures, but not their implementation or extra-judicial repression. To date, scholarship on the backlash against civil society has focused primarily on the adoption of restrictive legal measures. This is understandable, given that legislation is often publicly accessible and is a clear, visible manifestation of repression. But merely adopting new seemingly restrictive legislation does not necessarily mean that government is repressing civil society; this depends on the prior legal framework, as well as on a given law’s actual implementation and enforcement. It may be that restrictions are enforced, and NGOs disband and cease their activities as a result of that. But it may also be that legislation has been passed, but not implemented or enforced, and yet there is a measurable effect on NGO populations, because the legislation’s passing has a scare tactic effect. For instance, the adoption of a new restrictive regulation may signal the future ability of government to actively clamp down on NGO employees, or it may invite other actors to restrict NGO activities, or even engage in repression themselves (Fransen et al., 2021; Pousadela & Perera, 2021).

Scholars have begun to tackle this issue of measurement. Chaudhry’s (2022) measure of crackdown provides one example of more comprehensive measure of government repression; she uses data from Amnesty International for the years 1990 to 2013 to explain when states engage in violent versus administrative crackdowns against civil society organizations. The Varieties of Democracy (V-Dem) datasetFootnote 5 also provides longitudinal data based on expert opinion about government repression of civil society organizations. VDEM includes measures of government control over entry and exit of organizations; government harassment of, and relationships with, civil society groups; how citizens engage in organizations; and organizational independence and influence. However, as Chaudhry (2022) notes, the VDEM measure conflates legal and extra-legal repression (see her footnote 89). Though there are a few other attempts at cross-national coding of civil society restrictions, such as the CIVICUS Monitor of Civic Space,Footnote 6 these data sources do not cover a sufficiently long enough period of time to understand changes over time, and the concept they are coding (“civic space”) is poorly defined.Footnote 7 We must therefore also continue to engage in qualitative, in-depth case studies. Qualitative work will first be important to more fully understand what “repressive” actually means in a given context. Secondly, it will aid process-tracing to examine the operation of causal mechanisms that may or may not include a range of different variables.

A second limitation to our analysis is that we do not know what the implications of a reduction in organizational numbers in fact means, substantively. That is, we need a better understanding of what kind of weakening of civil society is taking place: whether a disappearance of organizations and fewer of their activities, or the informalization and de-politicization of civil society activity and the rise of more obscured, secret activities that necessarily are more complex to organize and likely less broadly societally impactful as a result. In response to a legal clampdown, NGOs may choose to refer to themselves as not being an NGO, according to the latest national legal definition, and thus not submit information to the Yearbook. If restrictive laws indeed have this impact on self-definition, the Yearbook misses data on NGOs that have stopped referring to themselves as NGOs, but whose activities may possibly still fall within the civil society category. Country case studies indicate that it is likely that legislation can push such an organization to re-brand itself as a “local” group or as a service-oriented organization, as occurred in Ethiopia when the 2009 Charities and Societies Proclamation was implemented (Dupuy et al., 2015), or may inspire NGOs to rebrand as social entrepreneurs or consultants (Fransen et al., 2021). More generally, qualitative studies indicate that the effect of NGO restrictions may result in both NGOs disbanding, as well as NGOs continuing in a more informalized, more precarious, and under the radar-mode with their civil society activities (Toepler et al., 2020). This too may lead them to be less visible for the Yearbook in terms of tracking their activities online and lead them to be unresponsive to Yearbook staff for updates on their activities. While this may leave the measurement of INGO numbers in analyses like ours open to debate about interpretation, the substantive message remains the same irrespective of what interpretation of numbers is taken when INGO numbers do drop. After all, both actual organizational disappearance and informalization imply a weakening of human rights INGO activity. Here too, further qualitative work may unearth dynamics more specifically.