Partnership in Times of COVID-19: Government and Civil Society in Austria

: How did the relation between Civil Society Organizations (CSOs) and government develop during the first months of the COVID-19 pandemic, once governments had taken restrictive measures to lock down economic and public life? Austria is used as an example of a corporatist welfare state whereby collaboration between government and CSOs occurs particularly in the ﬁ elds of social services, health-care, and youth. Our analysis focuses on the social service sector and differentiates between several social policy ﬁ elds. We hereby analyze data from qualitative interviews with CEOs from 30 CSOs, four group discussions with another 30 representatives of CSOs, public agencies and authorities, and from a standardized questionnaire (n = 99 CSOs). We also utilize our own experiences as participant observers in meetings between CSOs and government. Results indicate that CSOs suffered ﬁ nancially partly due to a decrease in income, though mostly due to an increase in cost. In social services in particular, they also faced hardships caused by the need to reorganize operations and human resources, and by the increased demands of customers. Federal government took responsibility for supporting CSOs ﬁ nancially, though such support was hampered by unclear competencies in Austria ’ s multilayered federal system. actors in this framework. In this study, we have outlined the financial situation, operational and client-oriented aspects, and the public-CSO collaborations as impact areas. In all of them, the fabric of the corporatist welfare system has met the challenges until now, but zones of specific vulnerability have become visible.


Introduction
On February 25th, two Italian citizens were the first official COVID-19 cases in Austria. The next day, the first Austrian citizen tested positive for COVID-19 in a Vienna hospital, where he had already spent 10 days in treatment. On March 12th, the first Austrian died from COVID-19. Simultaneously, the pandemic exploded in Ischgl, a ski-resort infamous for excessive après-ski partying. This hotspot, together with the media coverage from Lombardy showing overcrowded hospitals, mortuaries, and locked-down cities, triggered a swift and radical response from the Austrian government. On March 13th, the Austrian government enacted a nationwide lockdown for all schools, universities, kindergartens, all shops except groceries and pharmacies, and all restaurants and bars. On March 15th, the Austrian parliament enacted the legal foundation for future measures, including a stay-home-order for a month. On March 17th, all Austrian airports closed. Although the highstreets were empty and unemployment numbers were skyrocketing, the radical measures proved effective in holding back the initial spread of the virus. With only 85 deaths per million inhabitants, Austria did far better than many other countries during this first wave, e.g. the UK (626 deaths per million), the US (591), France (459), Switzerland (237), and Germany (112). 1 By the end of summer, the swift and consequent actions of the Austrian government had both prevented the health care system from being overburdened and curbed an excessive death rate. Yet, the measures caused substantive damage to the economy and wider society.
Civil Society Organizations (CSOs) of all kinds were severely affected by the lockdown: CSOs in fields of sports, arts, culture, and education suffered extremely from the prohibition of all events and the stay-home-order. Social and health care services faced enormous organizational challenges and additional costs. For some CSOs the demand for services either persisted or increased with the onset of COVID-19. Some vulnerable clients needed particular support, and some operations required complete reorganization. Social businesses operating in industries like tourism, gastronomy and events were specifically hit by the lockdown. The same was true for CSOs in youth work and education. As many parts of Austrian civil society are in close relation with different levels of government, our study will focus on how this relationship has developed during the COVID-19 crisis. We thus aim to address the following questions: How were CSOs affected by the COVID-19 crisis, particularly in the field of social and health care services, and how did the configuration of relations between CSOs and government develop in the first months of the crisis?
Austria's nonprofit sector has been described as showing strong elements of a corporatist nonprofit regime (Neumayr et al. 2017). Corporatist elements have become even more dominant as social services provided by CSOs have become increasingly important, e.g. in elderly care and childcare. Likewise, collaboration between the government, CSOs, and public funding is quite prevalent in arts and culture, sports, and education. In social services, almost 80% of CSO funding stems from public sources (Pennerstorfer, Schneider, and Reitzinger 2015). We will therefore concentrate our analysis on this field, referred to as social services in ICNPO, embracing the fields of human health, residential care, and social work activities in national statistics (NACE). Working within these fields are more than 3000 CSOs (30% of all Austrian CSOs with paid employees), together employing almost 150,000 employees (65% of all employees in Austrian CSOs), and yielding a 65% of the total 2.3% contribution of all CSOs to GDP (Neumayr et al. 2017). This makes social services by far the most important field in economic terms. There are also other important fields that also have close relations with government. Sports and leisure, for instance, are organized in a matrix of umbrella organizations that distribute public money to the individual associations. Even in sports, 16% of the income comes from government sources, with the federal government alone spending ∼€120 million per year. The funding situation in arts and culture is very similar. Austria's landmark public institutions such as the national theaters, opera houses, and museums, but also the many CSOs in this field depend heavily on public funding. In view of this, our analysis allows for a generalization over mainly formal, economically active Austrian nonprofit organizations with paid employees, though we will also further concentrate on social services.

Theory and the Austrian Context
The role of CSOs and their relationship to government varies between countries. Salamon and Anheier (1998) proposed a typology consisting of four different nonprofit regimes, formulated along two dimensions: government welfare spending and sector size. They labeled countries with a sizable nonprofit sector and high levels of welfare spending as either a corporatist or a welfare partnership regime. The social-democratic regime is characterized by high levels of welfare spending, but with a diminished role of nonprofit organizations in welfare provision. In the liberal nonprofit regime we find low levels of welfare spending, but a sizeable nonprofit sector. Finally, the statist regime has both low levels of welfare spending and a small nonprofit sector. In a similar vein, Young (2006) describes modes of government-nonprofit interaction. A complementary relationship is one of close cooperation, where both involved parties act as partners. Welfare services are funded by the state while nonprofit organizations have the capacities and knowhow to provide the services in many fields. This relationship is closely connected to Salamon and Anheier's (1998) welfare partnership type. In a supplementary relationship, nonprofit and public providers work in parallel, whereby the nonprofit organization fills a gap in state-run service provision. Finally, in an adversarial relationship, nonprofit organizations aim at influencing the public sector, demonstrating the advocacy function of the nonprofit sector (see also Neumayr and Pennerstorfer 2020).
While these theories describe the role of the nonprofit sector in relation to the government more generally, other theories focus on the contractual relationship of the two parties. Through the lenses of agency theory (Jensen and Meckling 1976), CSOs act as agents on behalf of the government (the principal) by providing welfare services. The theory describes the problem of incomplete contracts and the agent's opportunity to exploit these by making use of information asymmetries, such as a CSO knowing details regarding the success or failure of an ongoing project's operation. In order to overcome information asymmetries, public authorities seek to introduce monitoring mechanisms, incentives and sanctions as well as increased reporting requirements (Van Slyke 2007).
Another solution to cope with information asymmetries is relational contracting, which emphasizes long-term relationships, trust and reputation among the involved parties (Bertelli and Smith 2010;Brown and Troutt 2004;Christensen and Laegreid 2010;Isaksson, Blomqvist, and Winblad 2017). In this relationship, CSOs act as stewards (Van Slyke 2007). Goal convergence, shared collective interests, cooperation, joint decision making, and an exchange of information describe such a relation and point towards a more collaborative governance (see also Mendoza andVernis 2008 who further discuss the emergence of the relational state).
The relationship between government and CSOs in Austria can be categorized as a corporatist/welfare partnership regime. Traditionally, they play an important role in public service delivery in corporatist welfare states. Accordingly, Austrian nonprofit organizations deliver a major share of social services, with the relationship to the public falling under the welfare-partnership typology. Figures for the year 2013 show that the nonprofit sector accounted for 89% of total valueadded in the category "social work activities without accommodation", and almost 45% of total value-added in "residential care activities" (Leisch, Pennerstorfer, and Schneider 2016, 382).
This assessment is consistent with the more general interpretation of the Austrian state as a corporatist welfare state (see e.g. Österle and Heitzmann 2020). Federal, regional, and local authorities are all involved in welfare policy making and implementation. In general, the Austrian democracy is characterized as strongly consensus-oriented, and the so-called "social partnership" between employers' representatives and trade unions has a strong role in policymaking (Falkner and Leiber 2004). Additionally, the welfare state has a long history of continuity of policy making. Österle and Heitzmann (2020) describe recent changes within the Austrian welfare state in light of demographic changes, the 2008 economic crisis and the 2015 refugee crisis. They assess these changes as incremental and gradual rather than radical, and again highlight continuity as a key feature of Austrian welfare policy making.
These features also describe idiosyncrasies of the Austrian civil society, namely corporatism and federalism, as well as the close ties between nonprofit organizations and political parties and churches as defining elements for nonprofit organizations (Neumayr et al. 2017;Pennerstorfer, Schneider, and Badelt 2013). Nonprofits, especially large and well-established ones, enjoy a high level of political influence due to the country's system of social partnership. These nonprofit organizations often exhibit links to political parties or to the Catholic or Protestant church. Nonprofit organizations with ties to Christian-democratic ideas (relating to the People's Party or the church) or with ties to social-democratic ideas (relating to the Social Democratic Party) often exist side by side and deliver similar services (Neumayr et al. 2017).
Over the last 20 years, public authorities have increasingly sought to establish a quasi-market by introducing public tendering and refining contractual arrangements. This resulted in increasing competition among CSOs as service providers, and in some areas also to market-entry of for-profits (Heitzmann, Österle, and Pennerstorfer 2015). These new public management initiatives elicited increased professionalization and accountability among nonprofit organizations (Hwang and Powell 2009). Some authors describe marketization and hybridization of CSOs as signs of a change from a corporatist towards a post-corporatist welfare state (Bode 2011;Suykens, De Rynck, and Verschuere 2020).
These changes also become visible when examining the income structure of social service providers. In 2013, income from government accounted for almost 80% of total income of CSOs in social services (Pennerstorfer, Schneider, and Reitzinger 2015). Within this income category, one can further differentiate between income from contracts with public authorities (∼63% of total income in the category "social work activities except residential care") and income from public subsidies (∼17%). As in many other countries, the mechanism of public funding has shifted from subsidies to contracts (for an example, see e.g. Neumayr and Pennerstorfer 2020). Comparing the income structure in a panel of Austrian CSOs in 2005 and 2013 revealed that income from sales, which includes income from contracts with public authorities, increased by around 7% during this time period, while income from subsidies decreased by around 6% (Pennerstorfer, Schneider, and Reitzinger 2015, 41). The survey also revealed that public funding comes from a variety of sources at federal, regional and local levels. The regional level is the most important income source for many social service providers (Pennerstorfer, Schneider, and Reitzinger 2015, 21). On average, the organizations get funding from two different public authorities.
To estimate how Austrian CSOs dealt with prior crises, we turn first to the 2008 economic crisis. Simsa, Herndler, and Simic (2016, 19) report a reduction of public funding of social service providers due to the crisis, but CSOs were affected differently at the time and over the following years. Organizations often experienced funding decreases not so much in the form of direct cuts, but rather in the form of lacking inflation adjustments or through having to serve more clients with the same amount of funding. At the same time, CSOs judged their situation regarding public funding as rather good, and they reported a stable level of donations. Similarly, Pape et al. (2020) describe Austria's number of CSOs delivering social services as stable after the economic crisis. Pennerstorfer, Reitzinger, and Schneider (2020), however, reveal long-term effects by analyzing administrative data, finding that the 2008 crisis dampened sector growth. Yearly growth rates of the social services sector fell from 6.5% in the pre-crisis years to 3% in the years after the economic crisis. In the end, the number of service providers also shrunk, indicating market concentration. The decline in the sector's growth rates after the 2008 crisis can be explained with welfare state retrenchment.
Another more recent development with substantial implications for nonprofit organizations was the so-called European migrant crisis of 2015, whereby almost one million refugees came to or passed through Austria. Here, the role of CSOs was discussed not so much in the context of financing, but rather with respect to their capacity to react rapidly to the challenge. Meyer and Simsa (2018) highlight the crucial role of CSOs in handling the influx of refugees and their ability to fill the void left by public authorities. Some CSOs self-organized and worked with substantial input from volunteers to provide innovative services without public funding. The overextension of public authorities during this crisis altered the relation between the public sector and civil society (Simsa and Rothbauer 2016), bringing CSOs into a stronger position.
The 2020 COVID-19 crisis differs from the economic crisis of 2008 and the refugee crisis of 2015, amongst others, with respect to its impact on funding. The financial crisis caused less income cuts, but rather a long-term increase of demand for social services without an increase in funding provision due to public funding austerity. Similarly, the refugee crisis caused a tremendous increase of work for some CSOs in refugee care. In this case however, the government also provided funding to cover the costs of these measures. In contrast, the COVID-19 crisis led to an immediate income setback. At its core, it was and is a threat to public health and a challenge for the health care system and it triggered strict policies to contain the spread for the virus, which have caused enormous economic damage and collateral social damage.
Thereby, COVID-19 also presented numerous challenges to CSOs combining and going beyond the lack of funding or the necessity to expand service provision experienced in two aforementioned crises: As service providers in the health care system, they have played a central role in fighting the pandemic (e.g. the Austrian Red Cross and nonprofit hospitals). As care-providers and social workers, they had to protect vulnerable groups and employees (e.g. in clinics and residential elderly care, in social work centers dealing with homelessness and drug addiction issues). At the same time, due to the overall lockdown, they suffered financially from income losses and increased expenditures. In labor market integration, CSOs have been confronted with enormous demands as unemployment rocketed up from 3.8% in November 2019 to 6.2% in June 2020. The figure will most probably remain rather high, even in the unlikely event of a swift economic recovery (unlikely due to predictions of another oncoming wave of the pandemic). Therefore, the number of new challenges facing many social service CSOs are only likely to continue growing. Moreover, the pandemic simultaneously directly affected the CSOs' operations, reducing the ability to deliver services and disrupting organizations practices through the newly imposed contact restrictions.
In response to the current crisis, the Austrian federal government started to negotiate a fixed-cost subsidy for CSOs in the early stages of the pandemic, with the initial intention of complementing it with a fixed-cost subsidy granted to all business organizations. 2 One of the authors of this article participated as an expert advisor in these negotiations, which started in early April 2020 between the Federal Ministry for Arts, Culture, Civil Service, and Sports and the Federal Ministry of Finance. The first obstacle in the negotiations was to decide on the amount that should be dedicated to the nonprofit fund. As Austria still lacks a satellite account on nonprofits in its system of national accounting (United Nations 2003), the latest reliable data on the sector's revenue stem from 2013 (Leisch, Pennerstorfer, and Schneider 2016). For estimating the potential damage of the government's measures, these numbers were adjusted for inflation. The estimation of the potential COVID-19 damage yielded a likely loss of €700 million (see Table 1): For the field of arts and culture, losses estimated were validated by data on cultural events and visitors (Pitlik, Fritz, and Streicher 2020). For the field of sports, respective umbrella organizations contributed estimates of their members. For education, losses caused by the prohibition of seminars and trainings were estimated by the umbrella association of the continuing education providers. This estimate of €700 million was taken into account by the Ministry of Finance by the end of April. Subsequently, negotiations shifted to the federal Ministry of Tourism and Agriculture, which took the role of counterpart to the Federal Ministry for Arts, Culture, Civil Service, and Sports. It thus mirrored the two parties of the coalition government: the conservative People's Party and the environmentalist Green Party. This moment in April marked the beginning of some difficult work, starting with the initial selection and legal definition of beneficiaries of the fund. This turned out to be an extremely tricky task as there was no legal definition of nonprofits or CSOs in Austria. There is no specific legal form for CSOs, either: Associations and cooperatives are membership based and comprise the vast majority of organizations, yet public benefit corporations are establishing a foothold in social services in particular. There are also entities of public law and even entities of church law that are considered as significant parts of the CSO landscape. Eventually, it was decided that the statute in Austrian tax law that defines criteria for being charitable would serve as a proxy, without restricting the range of applicants to those CSOs who had already been granted the respective tax exemption.
After the initial selection and legal definition of beneficiaries of the fund, fixed costs had to be defined. Being restricted by both European competition law and by the regulation that was already enacted for business organizations, refundable costs did not include personnel costs. Instead, they comprised costs for rents, maintenance, additional costs for protecting staff and clients, and even support for keeping the infrastructure running, which was fixed at 7% of the revenue from 2019, i.e. any costs over 7% were not paid out. The overall payout per CSO was capped at the loss of revenues in 2020 in comparison to that of 2019, and at a total of €800,000 per CSO. From a comparative perspective, this nonprofit fund is more generous than the fund for business enterprises in Austria. The creation of the nonprofit fund marks a potentially significant change in the relationship between government and CSOs. Beyond its mere financial dimension, it is the first regulatory/legislative instrument that has legally acknowledged the nonprofit sector in Austria. It also expressed the federal government's responsibility for CSOs beyond the distribution of competencies in Austria's complicated federal system. Other remarkable changes, discussed in more detail in our findings section, were already visible before our data collection. These include the crucial role of the Austrian Red Cross during the whole crisis, the confusion of different regulations in Austrian provinces, the support of CSOs for the most vulnerable groups, e.g. in elderly care, and the remarkable lobbying power that umbrella organizations developed in the field of sports.

Methods and Data
We draw our findings on data yielded from 30 semi-structured interviews, an accompanying quantitative survey among CSOs in social service delivery and a set of four virtual group discussions hosted by one of the authors with participants from CSOs in the field of health-care and social services, public agencies and authorities. Hence, we applied a mixed-methods approach to yield insights into both the impact of the COVID-19 pandemic itself and the measures sought to fight the pandemic in the Austrian context.
Our qualitative interviews were conducted with 30 executives of Austrian social service CSOs, which we sampled according to fields of activities, number of employees, and geographical distribution. We focused on organizations providing social services in the fields of elderly care, housing and homelessness, violence prevention, disabled persons as well as children and youth. We covered organizations from all across Austria, namely the nine regional provinces, in order to obtain a comprehensive picture and to account for the fact that many fields of social service provision find their regulatory and public counterparts on a regional level.
In these semi-structured interviews, our questions covered a number of topics, including basic characteristics of the organization and the effects of the pandemic on organizational parameters such as funding, income generation, and human resources. Moreover, we investigated CSOs' responses to the imposed restrictions due to the pandemic on the delivery of services as well as the respective effects on the respective target groups. We also addressed how CSOs interacted with other organizations and with public authorities. Interviews were conducted via telephone (apart from one that took place in person) between the end of July and mid-August 2020. Each lasted 30-60 min. After each interview, the interviewer wrote short memos capturing the key statements. Subsequently, all interviews were recorded and transcribed verbatim, and content-analyzed.
In addition to these interviews, we refer to four group discussions that took place in late September and early October 2020. Each discussion lasted 120 min and included 7-8 representatives of large nonprofit and public institutions with expertise in public-third sector collaboration in Austria. We specifically asked participants to explain their experiences with the collaboration between public authorities, the public health care system, and CSOs active in the fields of health care and social services during the first months of the pandemic. The discussions were videotaped, and the results were clustered to identify the major strengths and weaknesses of these collaborations.
We complemented our qualitative data with an online survey. To this end, we gained access to participants via cooperation with an umbrella organization comprising 530 CSOs active in the field of social service and health care provision ("Sozialwirtschaft Österreich"), the members of which total 120,000 staff and around 200,000 volunteers. The rationale of the quantitative approach was to validate our findings from the interviews, generate numbers, compare these data with the business sector, and strengthen our empirical base. Therefore, we aligned the two approaches in content. The survey covered questions about organizational structure, fields of activities, and the effects of the pandemic on the organization as well as on the respective target groups (see also Appendix A). We also included items to measure CSOs' reactions and CSOs' financial reserves to also draw comparisons with surveys among Austrian business organizations conducted in April and May 2020 by the Austrian Institute for Economic research (WIFO, Hölzl, Klien, and Kügler 2020).
Our questionnaire was sent out on July 22nd and was accessible until August 7th 2020. In total, 167 organizations reacted to the questionnaire, resulting in 99 valid responses. The respondents represented 50,926 staff members, with 45,450 paid employees and 5476 volunteers. This yields an unweighted response rate of 19% and a weighted response rate (with respect to paid staff) of 42%, which is sufficient for the rather explorative and descriptive purpose of the study. On average, CSOs of our sample employed 459 persons (median: 37). Breaking down fields of activities, 52% of the responding CSOs are active in the field of social and health services (dominated by elderly care), 37% work with people with disabilities, 30% with children and youth, 15% on labor market integration, and 9% deal with housing and homelessness. Altogether, the surveyed CSOs cater to 223,006 clients (further descriptive statistics can be found in Appendix B).

Findings
CSOs in Austria have been affected by COVID-19 and the measures to fight the pandemic on various levels. We investigated CSOs providing social services for the elderly (e.g. residential and home care), for people with disabilities (e.g. training and occupation), for children and youth (e.g. childcare, kindergarten), for people concerned by domestic violence, and for the homeless. Figure 1 illustrates the degree by which organizations in these different fields reported being affected by the pandemic. Results reveal high levels of affectedness in all fields with a particularly large share (43%) of strongly affected organizations in the area of children and youth. In this field, many CSOs terminated all operations due to the stay-at-home order.
Respondents reported negative effects of the pandemic on different levels, namely on an organizational level (e.g. with respect to increased costs and sick staff or staff with care responsibilities), on the respective target groups (e.g. accessibility), and on the way that organizations have or have not interacted with Austrian governments (e.g. with respect to funding). As can be seen in Figure 2, volunteering in organizations was significantly affected due to restrictions of mobility and the practice of social distancing. It also shows that a majority of CSOs suffered from at least some form of revenue loss and cost increase, though both private donations and public subsidies remained widely stable.
In the remainder of this chapter, we will differentiate between (1) CSOs' financial losses and public support, (2) consequences for CSOs' operations, services, and target groups, (3) developments in the collaboration between government and CSOs.

Financial Losses and Public Support
Many respondents reported financial damage caused by the pandemic and the public countermeasures. CSOs faced increasing costs due to the provision of necessary protective equipment and training measures. At the same time, many experienced a simultaneous decrease in income. Though the latter was not so grave when compared to organizations in areas such as sports, arts and culture, a decrease in private donations threatened the financial health of many organizations, especially those that rely on personalized forms of fundraising. Among our survey respondents, 37% reported severe increases in personnel costs, caused by overtime hours and a Corona premium that was paid to all employees in caregiving. Almost 60% of the CSOs reported substantial extra costs for protective

equipment. Comparatively, the implementation of telework and digital solutions represented less of a strain on budgets, with only 14% of the CSOs reporting it to have been a large burden.
Half of the CSOs in the survey sample report a severe decrease in revenue. CSOs with a high share of earned income from markets and social enterprise business models experienced severe and existential decreases in income, while organizations with revenue from public sources (subsidies, performance contracts) remained less affected. However, some CSOs funded by performance contracts with governmental bodies suffered from their partners shutting down all operations, either federal and provincial agencies, or autonomous social security agencies. Even in cases in which costs were subsequently covered, delayed payments had an adverse effect on organizations' liquidity. In comparison, CSOs funded by subsidies reported that their financial situation remained largely stable. Hence, they were able to continue their work and balance additional costs with a reallocation of resources within their different sets of activities. Consequently, they did not apply for any financial support from the respective governmental crisis funds.
By end of September, more than 9000 CSOs had applied for financial support from the federal nonprofit fund, and the volume of the subsidies applied for exceeded €200 million. Almost €190 million of support has already been approved and almost €100 million had been paid out by end of September 2020 (see Table 2). Table 3 shows that the majority of CSOs that applied for financial support come from sports and recreation, whereas the biggest share of money has been allocated to CSOs in social services. In this field, the average amount of financial support paid out is above €142,000 per CSO. This corresponds with the fact that CSOs in social services are by far bigger than CSOs in other fields, which again is a result of the corporatist structure of the Austrian welfare state with the public authorities' preference to contract with a rather small number of bigger service providers.
CSOs applied different tactics to tackle their financial and operational problems (see Figure 3). During and after the lockdown, a majority of CSOs introduced and enforced telework. Furthermore, 59% of the larger CSOs and 41% of all respondents applied for the short-time-work program, which allowed employees to reduce their working hours without reducing salaries proportionally (e.g. 30% of

Partnership in Times of COVID-19
the working time, 80% of the salary). This program is funded by the public labor agency. Larger CSOs also used public securities and guarantees for liquidity loans, and almost a third of them postponed planned investments. Only 3% of our respondents had to lay off employees. Hence, CSOs with output-based funding relations (such as performance contracts) with the public sector reported the need for more security in funding. Across data sources, respondents called for more flexible funding schemes that are more closely tailored around the organizations' needs. In addition, several respondents pointed out that the uncertainty regarding which costs would be covered, and by whom they would be covered by, likewise imposed significant challenges on CSO executives and employees.
With respect to organizations' financial reserves, the dark sides of the corporatist welfare regime became visible. In the field of social service provision, CSOs as suppliers are often in an oligopoly with respect to the public sector as the dominant monopolistic payer. In the Austrian federal system, social services are essentially in the responsibility of provincial governments. They use their bargaining power for exerting pressure on prices, resulting in a revenue situation that does not allow CSOs to build up sufficient financial slack. As shown in Table 4, the majority of CSOs in our quantitative survey estimated that liquidity reserves even in the best case would last only for an additional six months. In the most likely case, around 70% predict that they would not survive longer than six months, and in the worst case 77% would run out of cash within half a year. Liquidity is much  better in Austrian business organizations: Surveyed with the same question, more than 50% of them report reserves that last longer than half a year, even in worst case, and almost a third of them do not expect any bottlenecks in liquidity (Hölzl, Klien, and Kügler 2020, 11).

Operations, Services, and Target Groups
CSOs were not only hit financially. In many cases the effects of the pandemic increased their workload, and the short-time-work program popular with businesses was much less helpful for CSOs. Where we see that more than 60% of all businesses have registered employees for the short-time-work program (Hölzl, Klien, and Kügler 2020, 7), only 40% of the CSOs of our sample did. In qualitative interviews, respondents reported that they often reorganized entire work units, or created new services or channels to be able to serve their beneficiaries while fulfilling the new safety measures required to tackle COVID-19. This increased workload during the pandemic distinguishes them from many companies in Austria. Social-service CSOs rather contributed to stabilization of the economy and the labor market with their comparatively smaller share of layoffs and short-timework. Still, 40% of CSOs registering employees for the short-time-work program is a remarkable share, resulting from the composition of social services that continued to full extent such as residential care, and others that were widely abandoned, such as social street work. We therefore also asked CSOs how strongly their clients have been affected by the pandemic (see Figure 4). Most obviously, clients were strongly affected by the restrictions of personal freedoms and the banning of visits to relatives, e.g. in residential care. In a similar vein, CSOs are very worried about threats to the mental health of their clients, also resulting from restrictions in personal communication and by the fact that many patients with mental disorders have somehow gone into  hiding, terminating their treatment. Likewise, restrictions of services, such as group counseling or joint activities have been massively detrimental to the clients. Due to the lockdown, access to the public space was drastically restricted, and care homes, day centers, shop floors and salesrooms, etc. were closed, in turn cutting down the social contact of all target groups. This was aggravated by the shutdown of public infrastructure relevant for CSOs to deliver their services effectively (such as canceled court hearings, or the closure of playgrounds). Being among the most vulnerable in society, these restrictions have put all sorts of stress not only on the beneficiaries and clients, but also to the CSO employees working with them. Not only did the restrictions impose many forms of stress onto the most vulnerable in society, but also indirectly onto the CSO employees working with these people. Daily structures and routines, key ingredients in social work, had suddenly dissolved. If at all possible, their restricted capability to cope with and adapt to new technologies created non-acceptance among the target group in many cases; e.g. people with disabilities or those suffering from dementia in the case of wearing masks. Group therapies and services could not be carried out any more due to physical distancing requirements. As personal contact is important for an effective and humane delivery of social services to the target groups, CSOs had to cope with an additional layer of psychological stress during the lockdown and the following period.
To summarize, social service CSOs of any kind faced specific hardships due to the lockdown's impact on their clients and the need for them to restructure their operations completely. In particular, in elderly care, protective measures for clients and staff, as well as the closure of residential institutions to visitors were most striking. For CSOs working with youth and children, the stay-at-home order and the school-closure combined with home schooling terminated most operations, and have deepened the educational divide. For CSOs dealing with domestic violence, the lockdown brought more emergency cases but with less opportunities for CSOs to make successful interventions. Work integration CSOs face increased demands caused by the rocketing unemployment rate.

Collaboration Between CSOs and Government
In our quantitative survey, almost 50% of the respondents report satisfaction with the support measures launched by the government, another 35% report neutrality. Only in cases where CSOs work with children and youth is the dissatisfaction rate significantly higher (50%; see Figure 5). This corresponds with the observation amongst all policy fields that education from kindergarten to high-school (K-12) performed worse during the pandemic, and CSOs received little support from public authorities. Leaving K-12 aside, the analysis of the collaboration between CSOs and government yields multifaceted results. It goes without saying that the time pressure to react to the spread of the pandemic has put a huge burden on the public sector. In addition to the numerous new work demands created through the pandemic and the countermeasures, CSOs had to deal with a high level of ambiguity. Many CSOs reported that they had wished for stricter and clearer safety regulations instead of mere recommendations for how to deal with staff and clients. The collaboration with the public sector was difficult, especially at the beginning, and sometimes resulted in a call for more centralized coordination among Austrian CSOs. Hence, regulations would have needed more details and stability, instead of rapidly implemented governmental enactments. CSOs also demanded better communication from the side of public sector about measures taken. Our group discussions with nonprofit and public sector representatives in late September and early October 2020 shed further light on the public-third sector collaboration during the pandemic. The discussions focused on the strengths and weaknesses of the collaboration between various levels of government and the CSOs active in social and health care services and revealed three overarching themes: First, collaboration worked best within frameworks that have already existed in some form before the COVID-19 outbreak, and where trust between the partners had been established before. For instance, provincial and federal governments each established emergency task forces in which CSOs were invited to collaborate from the beginning on, with an increasing number of CSOs involved. At the beginning, primarily the Austrian Red Cross participated in the major task force. Over the course of the lockdown, however, specialized task forces and crisis teams were established at different regional levels and for different fields. The more trust had already been built up in former cooperation, the better these new collaborations worked. In the City of Vienna, for instance, the umbrella organization for social service providers together with the Vienna public health care authorities managed procurement and distribution of protective equipment. In any case, the CSOs' and public authorities' welfare and common good orientation proved as a strong underpinning for the emergence of trust in the collaboration.
Second, rapid digitalization and switching to digital forms of communication was perceived as a strong enabling technology for collaboration, apart from the case of telemedicine, where it was also perceived as a result of good collaboration between public authorities, health-care providers, and caregivers. The swift implementation of new technologies brought on by the pandemic has accelerated processes and coordination between government and CSOs.
Third, in spite of pre-existing trust and merits of digital collaboration, collaboration turned out to be bumpy in many cases. One frequently mentioned challenge was the confusion caused by the multiple layers of Austrian federalism. In many fields, standards and regulations were either lacking or different between provinces, thus causing confusion in CSOs. For instance, the government decided to pay a Corona premium for employees in various social services, as they had to continue operations throughout the period of total lockdown when others were required to stay at home. Most CSOs paid this premium to the employees in July and applied for reimbursement, yet it was unclear, whether provincial governments were responsible for this refund (as they were the main funders of these CSOs) or the federal nonprofit fund would cover the cost (as the federal government had announced the premium).
In Austrian governments' policies to fight the pandemic, a single CSO played a pivotal role: the Austrian Red Cross. With 133 local units, 9000 staff members, 75,000 volunteers and more than a million members and supporters, the Austrian Red Cross is possibly the most powerful CSO in the country. From the early days of the pandemic on, the Red Cross played a decisive role in the emergency task forces at all levels. It took responsibility for gathering the PCR tests, it collaborated with the government in launching an information campaign on the pandemic and it developed and offered a smartphone app for contact tracing. In the course of our qualitative inquiry, other CSOs complained about a lack of integration in emergency task forces on federal level in the early weeks in spring, and that they were not informed sufficiently.
Most emergency task forces were formed either among CSOs and peers in their respective fields, or with the participation of provincial governmental bodies. In particular, CSOs focusing on specific target groups and smaller organizations wished to have been more involved in respective task forces and governmental decision making based on their to specific expertise with regard to clients and client work. It was generally reported that a deeper involvement of the wider social service sector in governmental decisions would have been appreciated, rather than what was perceived as the dominance of individual, usually larger, CSOs.
The federalist governmental structure of Austria regulates the competencies of providing social services in a rather complicated way. This also resulted in double and multiple burdens for CSOs. The different governmental frameworks across the nine Austrian regions and at the federal level sometimes created confusion and uncertainty, not only in terms of funding, but also in terms COVID-19 related rules and regulations that had to be dealt with. General enactments on the federal level had to be further specified and carried out on the regional level. Consequently, some provincial governments seem to have been more pragmatic and flexible in their handling of their partnerships with social service providers.

Discussion
Generally speaking, Austria's civil society has a partnership and a cooperative relationship with government, which is coined by a corporatist welfare regime. The COVID-19 pandemic has massively challenged all institutional actors in this framework. In this study, we have outlined the financial situation, operational and client-oriented aspects, and the public-CSO collaborations as impact areas. In all of them, the fabric of the corporatist welfare system has met the challenges until now, but zones of specific vulnerability have become visible.
At the same time, CSOs in the field of social service provision have also acted as stabilizers of the economy. Compared to for-profit firms, they were considerably less frequently laying off staff or using short-time work, once again emphasizing the importance of social services for the society and economy. The government has also recognized this significant role of CSOs in providing stable services, which is indicated by the quick establishment of funding support for them. This nonprofit fund partly helped reducing dependency on other public funders, at least in a short-term perspective. More precisely, the support from the federal government decreased dependence on contracts with provincial government bodies.
Due to their low levels of financial reserves, CSOs in social services and healthcare are especially vulnerable to economic shocks such as that, which resulted from the lockdown period. Constant under-funding of social services as a result of limited cost of living adjustments by the public sector has reduced the organizations' ability to bridge short-term revenue losses. CSOs that are mainly funded by performance contracts with the public authorities were often hit especially hard by the lockdown since they were not able to deliver (and charge for) their services. Organizations with higher levels of grant funding mentioned severe financial losses less often. Hence, with the ongoing increase in service contracting combined with only modest funding adjustments, CSOs in Austria have been made aware of the pressure to become more efficient in delivering their social services over the past years. This development has reduced the organizational capacity of CSOs to absorb external shocks such as the pandemic or other crises. Similar results can be observed for the issue of innovation in CSOs, whereby the absence of slack resources can have negative effects on the capability to come up with new processes or services (Meyer and Leitner 2018). This suggests reconsidering the financial aspect of the collaboration. More slack or a higher level of financial reserves might help to cope better with unforeseeable circumstances and reduce the resource dependence on public budgets, which are overstrained anyway in times of crises. This also relates to the observation that the larger the CSO, the better they are currently able to cope with crises financially. They show more flexibility in their ability to reallocate funds, often based on their variety of funding sources, and personnel between different units that have been hit by the restrictions in different ways.
Many CSOs, in particular smaller ones that focus on very specific areas, have developed a high degree of resource dependence from public authorities during the COVID-19 crisis. They were reactive rather than active, waiting for government help and action. There is a tradition of CSOs being on the short leash of governmental funding. Hence, the partnership has turned from being on equal terms to a rather paternalistic relation. From a short-term and rather cynical power perspective, public authorities might benefit from these asymmetries, as they might soften critical advocacy and make CSOs compliant and submissive. Insofar, the corporatist nonprofit regime has generated symptoms of paternalism. CSOs won't "bite the hand that feeds them" (Almog- Bar and Schmid 2014;Neumayr, Meyer, and Schneider 2015). In the long run, however, this dependence and the lack of organization slack will not only strain public budgets in times of crises, it will also turn CSOs into less innovative and effective agents in the implementation of public welfare policies.
Beyond the resource dependence of many CSOs, a rather modest level of ambiguity tolerance appeared as a second zone of vulnerability. After the first tough weeks of the state of emergency when national governance was dominated by order and obedience, regulations became more differentiated, revealing many conflicts and untouched and unregulated areas. Both in our interviews and in the group discussions, many respondents complained about these states of ambiguity, yet we found little reflection about the temporary inevitability of such situations when no one has routines available. Yet, the government did little to incentivize CSOs' entrepreneurial activities before. Again, for enhancing autonomous and flexible reactions on crises, government should reflect on the way that their funding schemes work, with particular regard to the trade-off between business income and grants, and how not to financially penalize CSOs that are successful in gaining market income and donations.
Of course, our study suffers from some limitations. In our analysis, we focused predominately on formal CSOs with paid staff in the field of social service and their relation with public authorities. Therefore, we do not know how grassroots and social movement organizations, neighborhood initiatives, and small CSOs are affected by the pandemic. Neither can we report about its impact on individual civic engagement, though preliminary results from a different data set of the authors on volunteering and private giving do not indicate major changes in this domain. Furthermore, it is important to point out that the study mainly described the first months of the pandemic in Austria, neglecting developments at later stages. Finally, both our qualitative and our quantitative sample may suffer from some bias. Whereas we did our best to ask CSOs that present the breadth of Austrian social service CSOs in our quantitative sample, it is difficult to assess the non-response bias in the quantitative sample.
What do we learn from the Austrian case? Austria serves as an example for a welfare partnership between public authorities and CSOs. The parties involved in this framework of formal and informal contracts have established a trustworthy relationship that goes beyond mere transactional collaboration. In times of crises, such a relational setting provides effective channels for trustful communication, facilitating swift response. This collaboration, however, tends to reinforce dependencies and paternalistic tendencies, which may also hinder CSOs from acting autonomously and creatively. 21. Which concrete socio-political measures do you recommend to the Austrian government to overcome the challenges posed by the COVID-19 pandemic?
Background 22. What legal form best describes your organization? (association, charitable private limited company, non-charitable organization, organization in/with strong affiliation to the public sector, other) 23. What is your role in the organization? (director/management/board, middle management, employee, other)