Trust in Institutions, Social Solidarity, and the Perception of Social Cohesion in the Republic of Moldova in the Early Phase of the COVID-19 Pandemic

Introduction

The COVID-19 pandemic is revealing of the evolution of trust in institutions and of social solidarity and cohesion throughout the world, and this article sheds light on the role of the key components of social cohesion during the pandemic within a society in which social trust and solidarity are both relatively low.

In the first months of the pandemic the Republic of Moldova, like other East European countries, reported a relatively low infection rate. It appears from COVID-19 Community Mobility Reports data presented by Google, which measured people’s mobility in public and residential spaces during the pandemic compared with pre-pandemic data, that Moldovans showed compliance with health regulations imposed by the government during the state of emergency from 17 March to 15 May 2020 (Spătari 2020). By way of explanation, experts have pointed out that precisely because of the weakness of the health system and expected vulnerability to the new virus, lockdown restrictions were firmly applied during the first days of the pandemic (Petrović et al. 2020). Later however, in the autumn of 2020, Moldova was among a number of countries in the region that had reported high rates of infection with the new type of coronavirus, and many deaths. On 2 December 2020 Moldova had registered 30,728 cases of COVID-19 infections and 655 deaths per million from a stable population of 3.5 million.^[1] By comparison, in neighbouring Romania the ratio was 25,259 cases of infection and 608 deaths per million, while Ukraine reported 17,382 cases of infection and 292 fatalities per million. Among post-Soviet countries only Armenia recorded more infections (46,275) and deaths (751) per million inhabitants. A significant number of Moldova’s population considered the official estimated exaggerated and denied the gravity of the crisis, and so neglected to observe the new public health rules. As anxiety over socioeconomic insecurity and distrust of state institutions grew, positive expressions of cohesion and solidarity and general observance of health rules gave way to tacit sabotage of government policies, especially those perceived to contradict the economic interests of the population such as social distancing rules and measures imposing restrictions on mobility (Negură, Gașper, and Potoroacă 2021).

Opinion polls conducted in Moldova beginning in May 2020 reveal low trust in government institutions.^[2] Confidence in the government fell sharply in 2015–2016 (from 23% in 2012 to 7% in 2015), following a case of large-scale bank fraud—known as “the theft of the century”—when it was discovered at the end of 2014 that a billion dollars disappeared from the accounts of three banks (Pilkington 2019). Low confidence in state institutions fuelled widespread distrust in the official interpretation of events during the early phase of the COVID-19 pandemic. According to a poll conducted in May 2020 by Watch-Dog and CBS-Research, half the respondents agreed that “the [COVID-19] virus is no more dangerous than common flu and everything is being done intentionally to destroy the economy”. President Igor Dodon himself downplayed the severity of the coronavirus in a YouTube programme in the pre-election period (Negură, Gașper, and Potoroacă 2021).

This paper examines a number of findings regarding social solidarity and perceived social cohesion, and trust in institutions in Moldova during the early phase of the COVID-19 pandemic. Many studies have addressed the relationship between certain components of social cohesion and the evolution of the COVID-19 infection and death rates, and we chose to measure trust in state institutions, a central component of social cohesion, in relation to people’s attitudes to COVID-19. In addition this study emphasises the dimension of socioeconomic insecurity in relation to trust in institutions and social solidarity and perceived cohesion. Qualitative analysis of in-depth interviews has highlighted other important aspects of social cohesion such as solidarity and the perception of social cohesion (or “social unity”, as referred to in the interviews) during the early phase of the pandemic.

Theoretical Framework: Social Cohesion, Trust, and Solidarity

For this article we applied the concept of social cohesion within the theory of social quality developed by W. Beck, Y. Berman, D. Phillips and others (Beck and van der Maesen 1999; Beck, van der Maesen, and Walker 2012). Using that model of social quality scholars define social cohesion as “the extent to which social relations, based on identities, values and norms, are shared” (Beck, van der Maesen, and Walker 2012, 61). Schiefer and van der Noll (2017, 592) define social cohesion as a “descriptive attribute of a community, which indicates the quality of collective coexistence”. That definition is consistent with the concept of Chan, To, and Chan (2006, 290) that

social cohesion is a state of affairs concerning both the vertical and the horizontal interactions among members of society as characterised by a set of attitudes and norms that includes trust, a sense of belonging and the willingness to participate and help, as well as their behavioural manifestations.

Drawing therefore on that model of social quality we focus in this article on various aspects of social cohesion such as trust in institutions, social solidarity, and perceived social cohesion.

Trust is a key aspect of social cohesion and social capital (Berman and Phillips 2012; Fukuyama 1995; Putnam 2000). According to a basic definition, trust is “the mutual confidence that no party will exploit another’s vulnerability” (Sabel 1993, 1133). In the social quality model, trust is a central component of social cohesion. As solidarity, trust belongs to the domain of “integrative norms of values” (Berman and Phillips 2012; Ward and Meyer 2009). A society is cohesive insofar as the members of that society trust each other and the society’s institutions (Chan, To, and Chan 2006; Schiefer and van der Noll 2017). Trust between people and institutions strengthens and harmonises the relationships in a community based on commonly shared norms (Fukuyama 1995). Students of social trust distinguish between “general trust” (Fukuyama 1995) or “generalised trust” (Hall 1999), which is a generic belief and attitude that helps people and communities to cooperate fairly and with mutual interest, and the “specific trust” between individuals and institutions (Berman and Phillips 2012). As it expands beyond the family, trust becomes more and more diffuse and anonymous (Fukuyama 1995). In the same vein, scholars distinguish between “vertical trust”—the trust in politicians and state institutions—and “horizontal trust”, shared among common people or peers (Sztompka 2006). Giddens and Luhmann delimit interpersonal trust, on the one hand, from trust in abstract systems. Institutional trust lies between those two poles and “radiuses of trust” (Fukuyama 1995; Giddens 1994; Luhmann 1979). Trust appears as a solution in situations of risk and uncertainty in relations between peers and between people and systems (Giddens 1994; Luhmann 1988; Sztompka 1999; Ward and Meyer 2009). To enhance people’s trust in government, and accordingly in institutions and policies run by the government, it is necessary to reduce the complexity of the government’s subsequent decisions (Luhmann 1979; Sibley et al. 2020; Ward and Meyer 2009).

Empirical evidence suggests that modern social developments in most democratic countries have led to the erosion of both trust of institutions and between individuals, whether those are democracies of long standing or post-authoritarian democracies (Dogan 2005; Sztompka 2006; Ward and Meyer 2009). According to scholars, the spread of corruption followed by repeated political scandals reported in media tend to fuel distrust both of politicians and governments (Dogan 2005; Montinola 2004; Sztompka 2006). The decline in trust might then turn the “climate” or “culture of trust” into a pervasive “culture of cynicism” which will affect social cohesion and societal wellbeing (Ward and Meyer 2009). Citizens might have “good reasons” not to trust politicians fully, whereas some forms and degrees of “institutionalised distrust” are bound up in the very principle of democratic politics (Dogan 2005; Sztompka 2006). Distrust becomes disruptive at the societal level when it is applied to specific persons or institutions and then moves into widespread systemic distrust—in science, or in democracy (Sztompka 1999).

Solidarity is another key concept of the social quality model that this article focuses on. The article examines the degree and forms of solidarity and support that respondents showed and observed in their own and others’ behaviour in the first months of the COVID-19 pandemic. According to a basic definition, solidarity refers to acts or attitudes of support or help given to other people (de Beer and Koster 2009; Voicu et al. 2020). Like trust, solidarity is a component of the domain of “integrative norms and values” within social cohesion theory (Berman and Phillips 2012). Scholars define solidarity using a number of dimensions. Horizontal solidarity characterises the manifestations of support between people within the same group, while vertical solidarity refers to acts and attitudes of help and support between people from different groups (de Beer and Koster 2009). That element of the definition is consistent with Emile Durkheim’s theory of mechanical solidarity (within communities) and organic solidarity (in modern societies) (Durkheim 1893). De Beer and Koster (2009) identify several dimensions that determine the manifestations of solidarity: the extent of its reciprocity (one-sided vs. two-sided); the type of organisation (formal vs. informal); the degree to which it is voluntary (voluntary vs. compulsory); its scope (local, national or global) and its form (that is, in the form of time, money or in-kind). According to another definition, subjective feelings of solidarity imply a set of components such as “trust” (“the belief that the exchange partner will not exploit the actor”), “affective regard” (“positive feelings for, and evaluations of, the partner”), “social unity” (“perception of the relationship as a social unit, with actors united in purpose and interests”) and “feelings of commitment to the partner” (Molm, Collett, and Schaefer 2007, 227). Societies where actions of support based on indirect reciprocity are the rule (in which the person giving support does not expect a direct reward) display higher social solidarity, including stronger trust, greater affective regard, perception of greater social unity, and stronger feelings of commitment (Molm, Collett, and Schaefer 2007). Hechter (1987) defines solidarity as a form of compliance with a group’s rules.

This article examines institutional trust especially in the government, as it relates to various expressions of solidarity and the perception of social cohesion—the “social unity” referred to in the interviews. We discuss trust and solidarity as parts of the domain of “integrative norms and values”, within the social cohesion theory and the social quality model.

A Review of the Literature on Social Cohesion, Trust and Solidarity in the Early Phase of the Pandemic

The literature on social cohesion and social capital have noted the resurgence of the COVID-19 pandemic, with many studies suggesting that both infection and mortality rates have been higher in countries and regions with lower social cohesion and social capital (Bartscher 2020; Bian et al. 2020; Borgonovi and Andrieu 2020; Jean-Baptiste et al. 2020; Kokubun 2020; Makridis and Wu 2020; Pitas and Ehmer 2020; Varshney and Socher 2020).

According to certain studies, social trust, especially in institutions, helped prevent the spread of the virus, because compliance with health rules depends on the level of trust in decision-makers (Bargain and Aminjonov 2020; Brodeur, Grigoryeva, and Kattan 2021) so that confidence in government is therefore an important determinant of compliance with quarantine rules (Bargain and Aminjonov 2020). One study claims that since mid-March 2020 people’s mobility has been significantly lower in countries with greater confidence in government and decision-makers (Bargain and Aminjonov 2020). Historical case studies and recent research suggest that when people do not support government health policies they might be less willing to comply with them. Moreover, low trust in government influences public opposition to compulsory government health policies (Taylor-Clark et al. 2005).

Studies show that high social cohesion, defined by collective adherence to norms, trust and values, determines higher cooperation by the population with state authorities and civil society organisations in combating the pandemic (Kokubun 2020; Makridis and Wu 2020). The increase of trust in institutions is achieved over time, through coherent, diligent interventions of the government and other state institutions carried out in good faith (Khemani 2020). Impracticable, inapplicable, and ambiguous rules might well create effects opposite to those desired (Elcheroth and Drury 2020). In countries where the government failed to take firm health measures in March and April 2020, community cohesion played a positive role in preventing the spread of the virus (Borgonovi and Andrieu 2020; Makridis and Wu 2020; Tufekci 2020; Varshney and Socher 2020). By contrast, in authoritarian China, compliance with health rules relied more on political trust and less on interpersonal trust (Wu 2021). Based on a comparative study of different government responses to COVID-19, Petrović et al. (2020) suggest that East European countries, with lower trust in institutions, had implemented more stringent interventions to enforce physical distancing, which gave better results in the initial phase of the pandemic than in most of the Western countries with greater trust in government. The explanation for that is that countries with less trustworthy governments and weaker health systems quickly imposed lockdown restrictions that proved to be effective in the short term (Petrović et al. 2020).

Trust in health services and trust in general correlate positively with past acceptance of vaccination against the A(H1N1) virus in 2009 (Rönnerstrand 2013). Increased social trust and a robust social support network has also helped prevent the spread of SARS, Ebola, Zika and HIV-AIDS epidemics (Chuang et al. 2015; Lundgren 2016; Makridis and Wu 2020; Muriisa 2007). Besides institutional trust and interpersonal trust, other factors that affect compliance with health rules are socioeconomic status, social responsibility, and attitudes to the severity of the virus (Brodeur, Grigoryeva, and Kattan 2021).

According to studies, major crises, disasters, or cataclysms can all mobilise societies, strengthen the feeling of belonging to a group, maintain social ties, and create solidarity among people and thereby facilitate both medium and long-term cohesion (Elcheroth and Drury 2020). Social solidarity develops at times of crisis in societies where solidarity is normally low (Voicu et al. 2020) but manifestations of solidarity might diminish if a significant part of that society does not feel equally vulnerable and then does not respect the rules, or if crisis management either produces or deepens inequalities (Elcheroth and Drury 2020). A study carried out in New Zealand found that during the first weeks of the pandemic/lockdown people reported greater trust in science, politicians, and police, higher levels of patriotism, but also higher rates of mental distress than before the lockdown (Sibley et al. 2020). A longitudinal study conducted in England suggests that in June 2020 perceived cohesion had declined, especially in the most deprived neighbourhoods, in the sense of “talking-to-neighbours” and “neighbour-trust”, compared to before the pre-pandemic (Borkowska and Laurence 2021).

The studies discussed in this literature review suggest that societies with higher social cohesion, defined by social trust, solidarity, and perceived social cohesion, were more effective in both preventing the spread of the virus and in combating the negative effects of the pandemic. All the same, the COVID-19 pandemic exacerbated the vulnerability of deprived groups and communities.

In the following sections we shall examine how Moldovan society’s social cohesion responded to the first months of the pandemic. Our analysis of quantitative data on institutional trust and attitudes to the pandemic is followed by analysis of semi-structured interviews on social solidarity and perceived social cohesion.

A Note on Data and Methods

In May 2020 the authors of this study, together with colleagues from the Centre for Sociology and Social Psychology of the Institute for Legal Political and Sociological Research (Institutul de Cercetări Juridice, Politice și Sociologice, CSSP-ILPSR) conducted a qualitative survey of 95 in-depth interviews with people of higher status in their communities or society in general. The authors and their colleagues based the interviews on an interview guide developed around a number of thematic blocks drawing on a set of indicators of social cohesion, namely belonging, participation, trust, solidarity, integrative norms, and socioeconomic security (Berman and Phillips 2012). The interview guide also contained questions about attitudes, behaviour, and social transformation during the first phase of the COVID-19 pandemic. The authors analysed the transcribed interviews based on the “grounded theory” model, using NVivo-10 software (Bazeley and Jackson 2013; Corbin and Strauss 1990). They structured the sample according to gender, place of residence, socio-occupational status, age, and ethnicity of the respondents, without pursuing national representation. Participants voiced their agreement to take part in the research, with a commitment to the anonymity and confidentiality of the resulting data.

The authors can corroborate the qualitative findings with the results of the questionnaire survey conducted by the CSSP-ILPSR research team in July 2020 on a nationally representative sample of 1202 adults. A number of questions in that questionnaire referred to the COVID-19 pandemic, in particular to fears during it, while one question was about satisfaction with the government’s management of it. The authors used the Worldometer dataset to analyse the data on the evolution of COVID-19 deaths and infection rates (Worldometer 2020). To analyse trust in institutions the authors used the data of the Public Opinion Barometer (POB) conducted by the Institute of Public Policy (IPP) and CBS-Research in October 2020 on a nationally representative sample of 1224 adult respondents outside the Transnistrian region (Institute for Public Policy and CBS-Research 2020). Data regarding trust in government in European Union countries are from the Eurobarometer-93 dataset which contains results from surveys carried out in July–August 2020 (Eurobarometer 2020). Data from Ukraine on trust in government are from the survey conducted by the Razumkov Survey Centre in July 2020 with a nationally representative sample of 2022 respondents (Razumkov-Centre 2020).

The primary hypothesis that guided our research, inspired by the literature on the relationships among various components of social cohesion and matters of public health, was that belief in the existence of the COVID-19 virus and compliance with the health rules aimed at preventing its spread are positively correlated with the level of citizens’ trust in state institutions. Research suggests that the correlation between public “belief” in COVID-19 and trust in institutions is neither linear nor self-evident, with significant differences between Western countries and Central and East European (CEE) countries (Petrović et al. 2020). We tried to test the hypothesis by comparing the case of Moldova with those of other countries in the CEE region and Western Europe. The secondary hypothesis, complementing the first one, is that belief in the COVID-19 virus and compliance with health rules depends too on perceived socioeconomic security during the pandemic. Statistical analysis of the answers to the questionnaire validates the first hypothesis largely, and the second one, partially. Qualitative analysis of the interviews contextualises and gives nuance to our findings. A working hypothesis that accompanied the qualitative analysis is that expressions of solidarity and perceived social cohesion intensified during the first months of the pandemic under immediate threat of the virus, but weakened as perceived socioeconomic insecurity and distrust in state institutions increased.

Trust in the Government and the Evolution of the COVID-19 Pandemic: Quantitative Findings

In the following we examine the correlations between the statistical data on the COVID-19 pandemic in various European countries (European Union (EU) countries, the United Kingdom, Ukraine, and Moldova) on the one hand, looking specifically at the number of infected persons per million inhabitants, the number of deaths per million inhabitants and the number of tests applied per million; and on the other hand assessing the level of trust in national governments.

For our analysis we used the Spearman correlation coefficient (Table 1). We found no statistically significant link between the number of fatalities per million inhabitants and trust in national governments (ρ = −0.324, p = 0.092), although we did observe a negative correlation, of medium intensity, between the number of deaths per million inhabitants and the level of trust in the government: ρ = −0.409, p < 0.05. A high level of trust in government therefore correlates with fewer deaths (the higher the trust, the lower the number of deaths). We observed a positive correlation, of medium intensity and statistically significant, between the level of trust in the government and the number of tests performed per million inhabitants: ρ = 0.437, p < 0.05. The greater the trust in government, the greater the number of tests performed.

These findings are consistent with the results of studies suggesting a link between the level of trust in government and some of the indicators for the COVID-19 pandemic in various countries (Bargain and Aminjonov 2020; Brodeur, Grigoryeva, and Kattan 2021). The number of COVID-19 deaths per million inhabitants is lower in countries where trust in the government is higher. The quantitative and qualitative analyses of public satisfaction with the measures taken by government during the pandemic performed within this research suggest that there could be two sides to the relationship between trust in the government and indicators of success in combating the COVID-19 pandemic. On the one hand, success in combatting COVID-19 increases when most citizens trust the government and comply with its actions. Equally, the level of trust in the government is higher when it manages to prevent the spread of the virus and at least reduce the increase in the COVID-19 fatality rate. Figure 1 helps us visualise the results from a comparative perspective by showing two groups of countries presented according to the intensity of the correlation between those variables. The cases marked with a dark coloured circle represent countries with a relatively high level of trust in the government and fewer deaths from COVID-19. The cases shown within the light-grey circle are countries showing less trust in their governments and which reported more deaths from COVID-19. The LOESS curve too suggests the same trend. There are, of course, significant differences within each of the two conventionally marked groups. Certain cases, such as Belgium, Sweden, or Ukraine, stand as exceptions confirming the model. Belgium reported both confirmed and suspected COVID-19 deaths as of 31 March, which influenced official mortality data from COVID-19 (Desson et al. 2020). Sweden adopted a less restrictive strategy in the face of the pandemic, focusing on personal responsibility and protecting the elderly and vulnerable. However, that strategy could not prevent the deaths of many elderly people in institutions (Lindström 2020). Finally, Ukraine joined other East European countries that imposed firm restrictions in the first phase of the pandemic, to the dismay of its neighbours Russia and Belarus (Åslund 2020; Petrović et al. 2020). But those restrictions have not proved effective in the long run, as evidenced by growing infection rates by the end of 2020 (Worldometer 2020).

Figure 1:

Trust in national governments and the COVID-19 deaths per million inhabitants.

As well as trust in national governments we examined two control variables, population density and gross domestic product (GDP) per capita,^[3] both of which variables have been considered in other studies, to analyse whether and to what extent they correlate with infection rate and COVID-19 mortality (Makridis and Wu 2020). We found the correlation between GDP per capita (USD) and the number of infections per million inhabitants so weak as to be insignificant (ρ = 0.01, p = 0.958) and the same for the number of deaths (ρ = 0.024, p = 0.903). Instead, there is a significant, positive, medium-intensity correlation between population density and the number of cases registered per million inhabitants (ρ = 0.387, p < 0.05). Similarly, we observed a significant, positive, medium-intensity correlation between population density and the number of deaths per million (ρ = 0.381, p < 0.05). Those results suggest that in different countries population density might be linked to COVID-19 infections and deaths, so that population density might affect other factors germane to COVID-19 infection and mortality rates, including those of a social nature studied here.

Along with other studies, this analysis therefore suggests a link between trust in state institutions and management of the COVID-19 pandemic. Besides, population density too might be linked with infection and mortality rates for COVID-19.

COVID-19 Scepticism According to Trust in Government and Parliament, Socioeconomic Status and Level of Education of Respondents

According to the Public Opinion Barometer surveys conducted by the IPP and CBS-Research on a nationally representative sample of 1124 adult respondents, in the Republic of Moldova confidence in the government evolved from 23% in 2012 to 7% in 2015 and 23% in 2020 (Institute for Public Policy and CBS-Research 2020). Meanwhile confidence in parliament decreased from 19% in 2012 to 6% in 2015 and 2016 and grew to 16% in 2020. The drop in confidence in ruling institutions in 2015 and 2016 was determined by the “theft of the century” revealed at the end of 2014 and mentioned above (see Table 2).

The level of trust in institutions especially the government, parliament, and presidency in the Republic of Moldova is less than in Romania, where according to Eurobarometer data from July–August 2020 confidence in the government is 34% and in parliament 27%, and less than in other European Union countries where—again according to Eurobarometer data from July–August 2020—average confidence in the government was 40% and in parliament 36% (Eurobarometer 2020). However, trust in institutions in Moldova is comparable to that of other countries in the region: Ukraine, Georgia, and Armenia (Table 3), except for the sharp increase in trust in state institutions in Armenia in 2019 following the change of government there.

^[4]Although a number of studies show a certain relationship between trust in institutions and infection rates in various countries (Bargain and Aminjonov 2020; Brodeur, Grigoryeva, and Kattan 2021) our study of Moldova and a number of other countries in the region found no statistically significant association between the variables mentioned above. However, the dataset out below (Table 4) suggest that trust in institutions might indeed be a factor that might, along with other variables such as socioeconomic security or population density, be correlated with success in preventing the spread of COVID-19.

Here are the results of the correlation analysis of the answers to the question “In your opinion, does the COVID-19 pandemic in the Republic of Moldova exist or is it a myth?” from the Public Opinion Barometer by IPP and CBS-Research in October 2020 and trust in the government and parliament, socioeconomic status, and level of education (Institute for Public Policy and CBS-Research 2020).

To make the results more readable we merged answers to the question “In your opinion, does the COVID-19 pandemic in the Republic of Moldova exist or is it a myth?” from the categories “certainly exists” and “rather exists” into one category and the categories “rather a myth” and “certainly a myth” into a separate category. The category “I don’t know/No answer” (DK/NA) remained unchanged.

The correlation analysis of the answers indicating belief or not in the existence of the COVID-19 pandemic, trust in the government, trust in parliament, socioeconomic status, and education provided statistically significant results. Among respondents who trusted the government the proportion who considered that the pandemic existed was significantly higher: 80.8% compared to 64.8% of those who did not trust the government. Also, among those who did not trust the government more people were undecided about the existence of the pandemic. We found a similar situation with trust in parliament. For sociodemographic status and education we found that the higher their level of education the more respondents believed the pandemic was real and that as their level of education increased fewer people gave uncertain answers (DK/NA).

Our findings are therefore consistent with the perception of vulnerability expressed by the respondents within the same survey. The survey conducted in July 2020 by the CSSP-ILPSR shows high perception of socioeconomic insecurity among respondents: 63.4% said they were afraid or very afraid that the COVID-19 pandemic would mean they could not provide the necessities for their families, with 67.3% confessing they feared being unable to cover the cost of services, and 47.8% afraid they would lose their jobs. It is noteworthy that people with primary or secondary education were most fearful of being unable to provide for their families—72.7% as against 58.6% of respondents with higher education. At least in Moldova, scepticism about the severity or even the existence of COVID-19 could be explained by the distrust of many citizens in state institutions, but also as a coping mechanism against the pandemic’s effects, perceived as detrimental to their socioeconomic security.

These results confirm our secondary hypothesis that awareness or denial of the COVID-19 pandemic correlates with socioeconomic status, level of education, and trust in state governmental institutions. The lower an individual’s socioeconomic status and level of education, and the lower that individual’s level of trust in state institutions, the more inclined was that individual to deny the severity or even the very existence of the COVID-19 pandemic. Given that people of lower education share to a greater extent a feeling of socioeconomic insecurity, the denial of the pandemic appears to be a “maladaptive coping strategy” to the risk of socioeconomic insecurity and exclusion (Jaspal and Nerlich 2020). Besides, socially disadvantaged people and those employed in manual labour might have what Boudon (1990) called “good reasons” for not believing in the severity of the COVID-19: they felt themselves forced to choose between falling ill with the virus and endangering their own and their family’s socioeconomic security.

We structured the following analysis around two main thematic blocks identified in interviews on social cohesion in Moldova during the COVID-19 pandemic. Thus, the key topics discussed by respondents and analysed in this article are (1) the social support and solidarity during the pandemic, and (2) the perceived social cohesion against COVID-19.

Solidarity and Social Support during the Early Phase of the Pandemic: Qualitative Findings

During the in-depth interviews respondents spoke about various forms of solidarity and social support in which they and other people became involved during the pandemic. The questions in the interview guide on this topic were: (1) What is/was the social unity manifesting through? Give examples. (2) How do you personally participate in establishing and maintaining connections/unity in the community? (3) How characteristic is it for people in your environment to help other people when they need it? (4) How do you personally consider from whom you can expect concrete material help in case of need (money, work performed, equipment)?

We can divide the subtopics of that generic theme into three broad semantic categories: subtopics with generally positive meanings, negative subtopics, and those with neutral-ambivalent meanings (Table 5). As expected from such a topic, subtopics with positive significance have the most references (6 subtopics/73 references).

The distribution of references according to the general significance of the sources for this topic suggests that most respondents who spoke on this topic positively appreciated the forms of social support and solidarity in the pandemic (Table 6). However, some references have been assigned to subthemes of negative significance, in particular when respondents criticised the lack or insufficiency of social support and solidarity. Other subthemes include thematic references with neutral or ambivalent significance.

Respondents who addressed the generic theme “Support and solidarity [against] COVID-19” in various ways mentioned different support actions: moral, psychological, informational support (regarding COVID-19, sometimes associated with the call for compliance with health rules), provision of necessities; food, clothing, sanitary equipment, blood donation, gifts of money, food delivery services, sanitation services, the support provided to students in the distance learning process and, not last, medical support.

In this article “the sources” means the interviews, and “references” designates the excerpts from interviews assigned to a particular topic and/or subtopic. Help was given most often to the elderly (17 sources/20 references—from now on: 17s/20r), but also to other vulnerable people (the needy, people with disabilities, the homeless, large families). Many people (9s/11r) helped those close to them such as relations, family members, neighbours, or colleagues. Many gave individual support either in a “horizontal” relationship between close people, or “vertically”, between unknown people, or from different social backgrounds. Some support was collective, from groups of volunteers. Finally, respondents employed in administrative structures spoke of forms of institutionalised support provided by local or central public authorities through the town hall, social workers, or medical institutions. A number of respondents mentioned the Metropolitan Church or various churches as foci of support for the elderly and the needy.

In this context, one respondent considered that manifestations of “horizontal” solidarity between relatives, friends, or colleagues could not be seen as proof of genuine solidarity. Genuine solidarity would have had to be solidarity shown with people from outside an individual’s close circle—“vertical” solidarity in other words—which according to that respondent is uncommon in Moldovan society (M/36–45/Chis/mold/sup/NG). Each reference is coded according to the sociobiographical data of the interviewee, as shown in Table 7. Thus, B/36–45/urb/mold/sup/pub means that the interviewee was a 36–45 year old urban resident Moldovan (but not Chișinău) man, highly educated and employed in the public sector.

The subtheme entitled “Solidarity actions” accumulated the largest number of sources and references (20s/34r). Examples of solidarity given by respondents were donations from abroad—the support of the team of Romanian doctors was mentioned especially (M/56–65/rur/mold/sup/pub)—and the decision to employ medical students in hospitals (B/66–75/Chis/mold/sup/pub) as well as taxi drivers who offered their services to doctors free of charge (2s/2r) and restaurants offering food to the needy (F/56–65/rur/mold/med/pub).

Those who provided help explained their gesture in terms of Christian duty (M/36–45/rur/mold/sup/NG), saying they acted out of moral obligation to the elderly and disadvantaged (M/36–45/Chis/mold/sup/NG) or from the promptings of family or kinship.

One respondent considered that groups of volunteer activists helping the needy emerged because the state did not involve itself sufficiently or at all (M/36–45/Chis/mold/sup/NG). Neither the state nor even the opposition parties did anything to encourage action in solidarity and for mutual support. According to the same respondent the state should have adopted systemic support measures to help vulnerable people beyond the pandemic period but in fact, on the contrary, they adopted instead isolation measures that discouraged them (M/36–45/Chis/mold/sup/NG).

A number of respondents denounced the lack of solidarity in their own society or community (8s/11r). Lockdown had led to isolation and thus to the reduction of communication and expressions of support and solidarity (F/36–45/urb/mold/sup/priv). According to one respondent the pandemic had deepened social inequalities: “We are not equal in the face of the pandemic […]. We are not in the same boat, we are in the same ocean, but in different boats” (M/36–45/Chis/mold/sup/NG). Instead of helping them, many people harassed those with COVID-19 and discriminated against them (M/36–45/Chis/jew/sup/pub; M/36–45/Chis/mold/sup/NG).

A number of respondents said they considered the best form of solidarity is to observe health rules in order not to contribute to the spread of the virus, some of them confessing that they had not considered other forms of involvement (M/16–25/urb/gag/sup/NG). The scepticism and negligence of some would be evidence of a lack of solidarity with the elderly, the most vulnerable in this pandemic, perhaps betraying latent “ageism” (M/36–45/Chis/mold/sup/NG).

One respondent argued that more collectivist societies with more “horizontal” solidarity, such as Asian societies, were fighting the pandemic more effectively while Western societies, being more individualistic, were therefore more vulnerable to the pandemic (M/46–55/Chis/rus/sup/pub-priv).

Much of what was said was of positive significance, while other remarks were negative or ambivalent. As with the theme “Unity against COVID”, references to positive subthemes were of lesser substance and contained fewer facts. The fewer critical speeches often brought arguments and examples showing the weak manifestation of solidarity in the COVID-19 pandemic in Moldovan society. Finally, some respondents expressed the hope that the crisis would stimulate the leaders of civil society and the public decision-makers to strengthen social support and solidarity, instead of weakening them through short-term measures like self-isolation at home. Those respondents saw such measures as ethically problematic and unsustainable anyway. Within communities, they said, decision-makers should especially encourage horizontal solidarity to support the elderly and other vulnerable people, and vertical solidarity between different social groups.

Perceived Social Cohesion Against COVID-19: Other Qualitative Findings

Many interviewees addressed social unity/cohesion in the COVID-19 pandemic. Respondents generally referred to the subject in answers to the following questions: (1) In your opinion, how united is/was your local community (in the context of the COVID epidemic)? (2) By what is/was this unity manifested? Give examples. (3) How do you personally participate in establishing and maintaining connections/unity in your community? (see Table 8).

Generally positively significant subthemes within this generic theme were fewer (n = 2) than generally negatively significant ones (n = 5) and than neutral-ambivalent ones (n = 4). However, there were most references to positive subthemes (n = 72), compared to negative (n = 29) and neutral-ambivalent ones (n = 37). This overview of the distribution of subthemes by their general significance suggests that respondents to the in-depth interview assessed positively the level of community and society cohesion in the COVID-19 pandemic. However, a smaller number of negative thematic references show that a significant fraction of the respondents negatively assessed the unity of their community and/or society. Finally, the many references to subthemes with neutral-ambivalent significance suggest that this conjectural cohesion, manifested by observing the new sanitary rules, was placing limits on the interviewees’ interpretation (see Table 9).

The statements in the subtheme “Unity – compliance with the rules” follow two perspectives: (1) the perspective of the authorities, in the case of persons with public authority status, and (2) the perspective of citizens, by respondents without civil servant status. Persons in authority usually spoke about how the authorities implemented the health rules. From the citizens’ perspective, interviewees usually described and analysed how ordinary people were complying with the rules. In both cases, respondents interpreted compliance with health rules as evidence of unity/cohesion in the community or society. In both discursive perspectives, respondents presented those who were complying with health rules as loyal citizens and described as “irresponsible citizens” those who were not complying.

Some respondents understood the level of unity (cohesion) through expressions of solidarity with vulnerable groups (F/16–25/urb/gag/sup/NG; F/56–65/urb/mold/sup/pub).

According to some respondents, the awareness of a common danger strengthens the cohesion of the community and society in a situation of major crisis and against a “common enemy” (M/56–65/Chis/mold/sup/pub-priv; M/66–75/rur/mold/sup/neang; F/26–35/urb/mold/sup/NG; F/16–25/urb/gag/sup/NG).

A number of respondents noted that unity was greater in March 2020, at the beginning of the pandemic when general fear was greater. People then felt they were “in the same boat”, “on the same wavelength”. That unity diminished, however, in the months that followed, as shared fear gave way to widespread distrust (F/56–65/urb/bulg/sup/pub).

According to one respondent, unity had grown in circles of those with close ties, especially among families, relations, neighbours, and colleagues, meaning that unity had diminished at the level of society at large (M/66–75/Chis/mold/sup/pub).

A significant fraction of the respondents considered that society was showing a lack of cohesion in the context of the COVID pandemic (12s/16r): “People have isolated themselves, but isolation does not lead to unity” (F/36–45/urb/mold/sup/pub). People have become more selfish and individualistic, “every man for himself” (M/26–35/urb/mold/sup/priv). The danger of infection had led to “self-closure and reduced communication” and thus to reduced solidarity (F/36–45/urb/mold/sup/priv). The lack of unity revealed itself in individualistic attitudes of indifference to other people, according to the proverb “My house stands aside and I don’t care about anything” (M/26–35/rur/bulg/sup/NG). “There was one who ran away from the hospital in Italy and infected the whole plane. What kind of unity are we talking about if there is no responsibility?”, one respondent wanted to know (M/56–65/Chis/mold/sup/pub-priv).

“United, in whose name? How can [the community, the society] be united if everyone is locked in apartments”, a respondent asked rhetorically (F/56–65/Chis/rus/sup/self-empl.). Then, the same interviewee continued her line of thought: “United by fear? Maybe, but not for everyone”.

Other respondents explained the lack of unity by the inability of the authorities to rally society in the crisis (for example, M/56–65/Chis/mold/sup/pub and M/36–45/Chis/mold/sup/NG).

As reported by respondents, people most often manifested their indifference to others by not following the rules and thus endangering others. There were a number of reasons suggested why people were not following the rules, including lack of awareness and neglect of others, especially of vulnerable people. More than indifference, some respondents saw hatred and malice in the behaviour of those who were not complying with health rules (F/36–35/Chis/mold/sup/priv). Denial of the epidemic and distrust of the authorities were further evidence of lack of unity (M/16–25/urb/gag/sup/NG). One participant believes that the way some people understand freedom in the pandemic affects the freedom of others. Caring only about their own freedom, those who were not complying with the quarantine rules were extending the quarantine and isolation for everyone else (F/46–55/rur/mold/sup/pub).

Most respondents articulated their arguments based on an implicit assumption that complying with health rules was the definition of community/societal unity/cohesion. Thus, some respondents considered that unity was missing because of non-compliance with the quarantine rules (9s/10r). Also, many respondents (44s/71r) appreciated that the unity of their community had increased due to the observance of health rules.

One respondent noted that in the pandemic conditions the sanitary-epidemiological rules were what ensured cohesion, while under normal conditions—before the pandemic—the main community binder was social closeness, especially kinship (M/56–65/rur/mold/med/self-empl). One respondent noted high cohesion in terms of compliance with health rules, but lack of cohesion in terms of genuine forms of solidarity and mutual aid: “[The community] was quite united in organizing protection measures, it was quite aware of the risks, but [it was] quite unsuccessful in terms of empathy for sick people and awareness that they should not interact with sick people” (M/26–35/Chis/mold/sup). One respondent noted that community activism to ensure health conditions in residential buildings had increased (M/26–35/Chis/mold/sup).

There were more thematic references to subthemes with generally positive significance than to those with negative and neutral-ambivalent significance. However, remarks attributed to positive subthemes were often redundant and lacking sufficient argumentative elements, many of them drawing on the belief that “the community is/was united because people respected the health-epidemiological rules”. References to subtopics of generally negative and neutral-ambivalent significance covered a wide range of ideas and arguments, a plausible explanation for the differences being that statements attributed to positive subthemes more closely echoed the official discourse of “managing the pandemic” and less closely the private beliefs of respondents. However, it is noteworthy that most respondents subscribed to the implicit definition of social cohesion as respect for the rules imposed by the authorities. The context of the pandemic perhaps determined that element of their definition, where a common danger was seen as strengthening the normative binder and weakening cohesion stimulated by other principles, such as those of belonging, trust, and solidarity.

Discussion of Results and Conclusions

This mixed-method research was carried out in May–July 2020 as a team research project of the CSSP-ILPSR and captured a number of trends showing certain fundamental elements of social cohesion in the Republic of Moldova during the early period of the COVID-19 pandemic, in the following aspects: trust in institutions, social support and solidarity and perceived cohesion.

The results of the survey carried out in June 2020 by CSSP-ILPSR, corroborated by the data of the Public Opinion Barometer conducted by the Institute of Public Policy and CBS-Research in October 2020 (Institute for Public Policy and CBS-Research 2020), suggest that awareness of the existence and severity of the COVID-19 pandemic correlates positively with socioeconomic status, level of education and respondents’ trust in the state’s ruling institutions. The lower the perceived socioeconomic status of the respondents, the lower their education and the less their trust in institutions, and the greater their tendency to deny the severity of the COVID-19 pandemic. A comparative statistical analysis of EU countries, Ukraine, Moldova, and the United Kingdom suggests too that government confidence correlates negatively with the number of COVID-19 deaths per million inhabitants.

Participants in the in-depth interview highlighted and described many acts of solidarity especially for the benefit of the elderly and other socially vulnerable people, carried out in various communities and in society at large to prevent people from becoming ill and to mitigate the risk of socioeconomic deprivation resulting from the crisis. In formulating potential solutions, criticism too was quite freely expressed. A number of respondents were hoping that state authorities and civil society leaders would learn from this health and societal crisis to strengthen social cohesion. Several participants in the research suggested too that the government and civil society initiatives had distributed support unevenly in the crisis and that such support should be coordinated to include more of the elderly and disabled people who are the most vulnerable in the pandemic.

The interviews highlighted an interestingly significant aspect of social cohesion in the COVID-19 pandemic. Respondents defined the unity/cohesion of Moldovan society most of all by observance of health rules. Most references on the subject were to appreciate the high level of unity in society during the first months of the pandemic. For those respondents, quite apart from other cohesive factors such as belonging, trust, and solidarity, the high risk of infection and socioeconomic deprivation reinforced the importance of adherence to the dominant rules.

This paper confirms the relevance of studies that suggest a link between different components of social cohesion and how the authorities and society took part in combating the pandemic and its negative effects. Yet, the analysis of the Moldovan case brings some significant nuances to the general picture and contributes to a critical discussion of studies on social cohesion and social capital in the COVID-19 pandemic. The study shows a correlation between denial of the pandemic among those of certain sociodemographic categories, especially young people, those self-identified with lower social status and those with lower education. That finding suggests stratified distribution of the perception of health and socioeconomic danger caused by the pandemic. Young people with lower human capital feel most at risk from a socioeconomic perspective in the pandemic, which confirms also the critical analysis of the social capital literature that suggests social support and expressions of solidarity circulate unequally in society (Portes and Landolt 1996).

Many respondents to the qualitative research consider that Moldovan society showed high compliance with health regulations aimed at preventing infection with COVID-19 and that it developed various forms of solidarity, both horizontally within communities and vertically between various social groups. However, a few months after the beginning of the pandemic citizens reached a “fatigue-threshold” in the manifestation of social cohesion through collective observance of health rules, physical distancing, and social solidarity. As concerns about the danger of socioeconomic insecurity because of the pandemic increased, positive forms of cohesion increasingly began to give way to negative psycho-social manifestations of “maladaptive coping strategy” such as denying the severity of the pandemic (Jaspal and Nerlich 2020). Such responses were even more pronounced among those whose trust in state institutions was low. In the perception of several respondents Moldovan society has not adopted sufficient collective resilience strategies despite or beyond the state’s attempts to manage the crisis, such as those applied in certain Asian societies (Bian et al. 2020; Tufekci 2020).

This article contributes to the research on social cohesion in the early phase of the COVID-19 pandemic with a focus on Moldova, a society in which trust in institutions is low. The article provides evidence of the link between institutional trust and the degree of success in managing the pandemic in Moldova in comparison with other countries from the region and the EU. Alongside other studies that correlate various components of social cohesion or social capital with indicators of success in combating the pandemic, this study highlights the significance of social status in assessing the severity of the pandemic. The scepticism about the severity of the virus is greater among people who acknowledge their position in the lower strata of society and who express feelings of increased socioeconomic insecurity in the pandemic. Other studies too examine the relationship between components of social cohesion or social capital in the pandemic from a purely quantitative perspective, and in interviews with 95 respondents this article proposes a qualitative analysis of solidarity and perceived social cohesion. The interviews analyse various forms of solidarity but criticise the limits of social solidarity in the pandemic. The interviews emphasise the significance of compliance with health rules in maintaining and strengthening social cohesion.

Drawing on the results, we would recommend to state authorities and civil society leaders: (1) to transform pandemic management from an individualistic model, focused on the “stay at home” slogan, to a solidarity-based model, which would favour flows of social support to vulnerable strata and those whose vulnerability deepened in the lockdown; (2) to foster society’s resilience in time of crisis by strengthening the infrastructure for social participation and solidarity (social volunteering, donations for social projects, online activism, etc.); (3) for state institutions to intervene efficiently and synergistically in time of crisis by collaborating with community-based and societal support structures.

Our study is not without limitations. In our analysis of quantitative data for example, we aimed to measure the correlations between the variables considered. The next step ought to be a causal analysis of the data, but for that we should need additional data. In our study we decided not to use prediction models because insufficient data were available. For regression analysis it would be necessary to include cases from more countries. An analysis based on a prediction model remains a goal for the future, while another limitation concerns the nature and consistency of the research data. The COVID-19 data used in the analysis reflect the state of affairs in late 2020. For a broader causal analysis a longitudinal approach would be necessary. Furthermore, we considered that institutional trust and other variables could be more appropriately analysed through quantitative methods, while the perceptions of solidarity and social cohesion would be more suitable for qualitative research. We therefore did not pursue thematic consistency between the quantitative and the qualitative analysis, a choice determined by the data we produced based on a methodology (the interview guide and the questionnaire) developed in the early period of the pandemic when its evolution was still highly uncertain. Despite those limitations we believe our article contributes significantly to the understanding of a process that is still ongoing.