Does pro-sociality or trust better predict staying home behavior during the Covid-19?

Abstract

Voluntary compliance of preventive and mitigation measures due to social concerns can play a crucial role in slowing down the spread of the Covid-19. The existing economic models for disease spread however do not direct a lot of focus on the possible role of pro-social behavior and general trust in predicting preventive behaviors amid the Covid-19. Therefore, this study analyzes whether pro-sociality and general trust measured in the short run (2020 and 2019) and in the long run (2015 and 2010) predict attitudes towards the stay home behavior and the intended stay home behavior in case the government mandates it due to the Covid-19 in the Netherlands. The results suggest that these preferences positively influence attitudes towards staying home behavior. However, trust in comparison to pro-sociality is a stable and robust predictor of stay home attitudes both in the short as well as long run. On the other hand, neither trust nor pro-sociality influences the intended stay home behavior in case the government mandates the lockdown, and it is most likely due to the timing of the survey coinciding with a significant drop in the Covid-19 infections and easing out of the lockdown restrictions by the Dutch government.

1. Introduction

The Covid-19 has caused significant damage to the whole world, imposed unprecedented challenges for the medical personnel, general public and governments, and the socioeconomic disruptions continue to massively plague most countries. It has been largely established now that the Covid-19 spreads through droplets with a transmission rate varying from 2.24 to 3.58 (Zhao et al., 2020). Preventive strategies and mitigation measures that include but are not limited to staying home and avoiding public as well as crowded spaces, frequently washing hands and wearing mask have been touted as effective tools to mitigate the rapidly surging virus cases (Quadri, 2020).

The Covid-19 preventive measures can be effectively established and observed if people care about social repercussions of their actions or have ‘social preferences’. The economic literature frequently reports that people depict social preferences and care about externalities of their actions (for example: Andreoni & Miller, 2002; Bolton & Ockenfels, 2000; Fehr & Schmidt, 1999; Fisman, Jakiela, & Kariv, 2017 and others). The Covid-19 experience so far from countries like Japan and Sweden where most people followed the Covid-19 preventive and mitigation measures despite no strict lockdowns reflects the role of social preferences in health behaviors. Contrarily, several existing studies suggest that when people are uncertain about social impact of their own actions, they tend to use this uncertainty as a justification to behave selfishly (Exley, 2016; Haisley & Weber, 2010; Schweitzer & Hsee, 2002). Therefore, there is a possibility that selfish behavior might triumph other regarding preferences and people might not follow the Covid-19 preventive measures that can ultimately lead to a surge in virus infections and a strain on the health systems.

The health economics literature modeling the epidemiological diseases has rarely considered the effect of social preferences in predicting the health behaviors (Campos-Mercade, Meier, Schneider, & Wengström, 2021). However, post-pandemic there has been an active attempt to analyze whether social preferences can be used as an effective tool to predict and model health behaviors related to the prevention and mitigation of the Covid-19. For example, Campos-Mercade et al. (2021) find that pro-sociality predicts preventive behavior in Sweden, Bargain and Aminjonov (2020) report that pre-pandemic trust in policy makers influences social mobility in Europe, Müller and Rau (2021) find that risk and time preferences predict the Covid-19 preventive behaviors in Germany, Fang, Freyer, Ho, Chen, and Goette (2021) report a positive relation between pro-sociality and preventive measures in Germany and Barrios, Benmelech, Hochberg, Sapienza, and Zingales (2021) report that civic capital in both Europe and the US is linked to social distancing. Similarly, Cucciniello, Pin, Imre, Porumbescu, and Melegaro (2021) report positive influence of altruism on intentions to get vaccinated in Italy while Yu et al. (2021) find positive influence of pro-sociality on intentions to get vaccinated in China.1

In the current study I extend the aforementioned literature and examine whether prosocial behavior or general trust measured in the short run (around pandemic time) and long run (well before pandemic) is a stable and better predictor of attitudes towards staying home behavior and the intended staying home behavior in case the government mandates a lockdown during the Covid-19. The analysis is performed using data from the LISS panel administered by CentERdata (Tilburg University, Netherlands). The prosocial behavior is based on donations made to the social causes in last 12 months while the general trust is obtained from a standard trust question. The prosocial behavior measured in 2020 and 2019 are used to examine the short-term effect while prosocial behavior and trust of the same participants measured in 2015 and 2010 are used to examine the long-term effect of these preferences on staying home behavior. The staying home behavior is measured with the help of three items. The first two items (appropriateness of staying home in case of symptoms and in case of no symptoms) essentially measure the attitude towards the aptness of staying home behavior while the third item (would stay home if the government mandates it) captures the intended behavior. I primarily focus on the staying home behavior because it is one of the effective strategies to control the spread of virus.2 Also, people in the Netherlands have shown strong opposition to the lockdown policies and police force has been used to disperse large crowds protesting against these measures.3 The unwillingness of Dutch people to accept lockdown measures provides an important context to examine whether and how social preferences can be used to nudge the Dutch people to stay home, prevent the spread of virus and ease load on the health and economic systems. Due to these reasons, I focus on the staying home preventive measures in the current study.

Theoretically, prosocial people might care more about the repercussions of their actions on others and therefore follow the Covid-19 preventive and mitigation strategies to minimize negative externalities of their actions. Similarly, the observance of preventive measures can depend on the interpersonal trust (Algan, Cohen, Davoine, Foucault, & Stantcheva, 2021). Higher trust in others might lead an individual to have positive expectations that others will follow the stay home recommendations and reduce public crowding.4 As a result, citizens are likely to observe the norm of staying home when they expect others will also follow the norm (Scholly, Katz, Gascoigne, & Holck, 2005; Yuan, Long, Huang, Huang, & Luo, 2022). The existing research also suggests that trust plays a positive role in developing useful partnerships among communities to collectively respond to epidemics and to accept the efficacy of strategies designed to counter future emergencies (Nuriddin et al., 2018; Ryan et at., 2019; Yuan et al., 2022).

The analysis offers weak evidence that pro-sociality measured in the short-term (2019) and in the long-term (2015) predicts attitude towards the appropriateness of staying home in case of virus symptoms only. On the other hand, trust measured in the short-term (2020 and 2019) as well as in the long-term (2015 and 2010) is positively associated to the attitude towards the appropriateness of staying home in case of virus symptoms as well as in case of no virus symptoms. Neither pro-sociality nor trust measured in the short or long-run predict the intended staying home behavior in case the government mandates it. The absence of relation between preferences and the intended stay home behavior is primarily due to the survey timing that coincided with a significant drop in the Covid-19 cases and the easing out of lockdown restrictions in the Netherlands. Overall, the results offer relatively strong support that trust in comparison to pro-sociality is a stable and better predictor of attitude towards the voluntary staying home behavior in the Netherlands.

The study makes several important contributions to the literature. First and foremost, it uses pro-sociality and trust measured in 2015 and 2010 well before the Covid-19 to examine the long-term role of preferences in predicting health behaviors. To the best of knowledge, this is the first study of its kind to examine the actual long-term relationship between preferences and preventive behavior and subsequently unravels the possible application of social preferences in predicting health behaviors in the long run.

Second, the study reports a comparative analysis to identify whether pro-sociality or trust is a better predictor of health behavior. This comparative analysis provides enough evidence to suggest that trust is a better and stable predictor of health behaviors in the Netherlands and highlights the relative applicability of trust in the health economics domain. It is however important to note that the measurement tools for the prosocial behavior and general trust are not comparable; the former is based on donations in the last 12 months while the latter is elicited through a Likert scale. Also, prosocial behavior is indirectly measured through donations while trust is measured through a direct question. These differences in the operationalization of prosocial behavior and trust might influence their effectiveness in predicting the health behaviors and this is a shortcoming of the current study.

Third, as preferences can be influenced by the pandemic itself, the preference data collected post-pandemic might lead to unreliable estimates due to the endogeneity problem. Several studies that examined the possible role of social preferences in predicting the health behaviors suffer from this issue. To overcome this possible endogeneity, I use preferences data from same individuals obtained through identical survey questions in a post (2020) and pre-pandemic (2019) situation to examine the robustness of results. Campos-Mercade et al. (2021) also use a similar design to control for the possible endogeneity, but for pre-pandemic analysis, the authors use pro-sociality measured in 2018. As preferences can be influenced by personal and professional experiences (Carpenter & Seki, 2011; Chuang & Schechter, 2015) and hence their strength can change over time, preferences measured right before the pandemic are ideal to perform the current analysis because they minimize the time window for the effect of such experiences. Considering this concern, my design is more refined because I use preferences data obtained through same instruments in 2019 (pre-pandemic) and 2020 (post-pandemic).

Lastly, this study adds to the growing pandemic literature that examined the role of prosocial behavior and interpersonal or institutional trust to study the Covid-19 preventive and mitigation behaviors (for example: Algan et al., 2021; Bargain & Aminjonov, 2020; Barrios et al., 2021; Bicchieri et al., 2021; Brodeur, Grigoryeva, & Kattan, 2021; Durante, Guiso, & Gulino, 2021; Schmelz, 2021 and others).

The rest of the study is organized as follows. Section 2 reports data and statistical analysis details. Section 3 discusses the results while last section concludes the paper with relevant discussions.

2. Data and statistical analysis details

The study uses data from the Longitudinal Internet Studies for the Social Sciences (LISS) panel administered by CentERdata (Tilburg University, Netherlands).5 The panel is administered to a representative sample of about 5000 households from the Netherlands and the respondents are incentivized for their participation. The panel has several modules and these are administered over several months. The current study combines data from the ‘Background Variables’ collected in December 2020 to gather demographic, income, housing and work information that serve as control variables in the regression analysis. The information about staying home during the pandemic (dependent variables) is obtained from the special module ‘Effect of the Pandemic on Social Trust’ administered in July 2020. This module was administered to a smaller number of households (n = 2698) and 2397 households provided complete response. The primary explanatory variable prosocial behavior and general trust are obtained from the survey modules ‘Social Integration and Leisure’ administered in October-November 2020 and ‘Personality’ administered in May-June 2020 respectively.

Except the special module ‘Effect of the Pandemic on Social Trust’, all other survey modules are annually administered from 2008 onwards. Therefore, information about prosocial behavior and trust can be traced back until 2008 and importantly, this information is obtained using same survey questions over the years. For the short-term effect of preferences on staying home behavior, the information collected in 2020 is used. As a robustness check and to control for possible endogeneity, I use data from 2019 for the respondents who participated in 2020 pandemic survey as well. There is not a big difference in 2020 sample and 2020 – 2019 matched sample. Depending on the stay home measure, 2020 – 2019 matched sample is smaller by only 38 to 43 observations. For the long-term analysis, the data collected in 2015 and 2010 from the same respondents who also participated in 2019–2020 surveys is used and these years are selected to keep a gap of reasonable time from the pandemic.

Before proceeding further, it is useful to have a quick look at the progression and unfolding of the Covid-19 in the Netherlands. Table 1 briefly summarizes the chain of important events and government policies to contain the spread of virus in the Netherlands.6 The timeline is restricted till July because the Covid-19 related survey was implemented in July. The first Covid-19 case was detected on 27th February 2020 and in less than a week's time the cases rose to 82 on 5th March. From there onwards, the cases and number of deaths kept on rising and in less than a month's time the total Covid-19 related deaths surpassed 700 mark on 27th March.

Table 1.

Timeline of Covid-19 in the Netherlands.

Dates	Events / Government Policies
27th Feb, 2020	The first Covid-19 case detected in the Netherlands. The 56 years old person travelled from the Northern Italy.
5th March	The confirmed Covid-19 Dutch cases increased to 82.
6th March	The first Covid-19 related death reported in the Netherlands.
9th March	The Dutch prime minister advises against handshaking and encourages people in the Noord-Brabant region to work from home.
10th March	The Dutch government advised to travel to Italy only for essential purposes.
12th March	The advice to work from home is extended to all regions in the Netherland. Social gatherings of more than 100 people are banned.
15th March	Intelligent lockdown imposed and schools, restaurants, bars, cafes, sports and leisure clubs closed and people instructed to observe social distancing. The number of Covid-19 cases surpass 1000 mark.
23rd March	Strict lockdown and monetary fines introduced for people and businesses not complying with orders. Social gatherings reduced to 3 people.
27th March	Covid-19 cases surpass 10,000 mark with 771 total deaths due to the virus.
31st March	Lockdown extended till 28th April. 175 people die due to the virus.
7th April	A record 234 deaths in a day due to virus are reported in the Netherlands.
21st April	The government announces to ease down the intelligent lockdown measures and allow primary school students to reopen from 11th May.
3rd May	The death toll surpasses 5000 mark.
6th May	The decision is made to mandate masks on public transport from 1st June.
11th May	Primary schools, day care centers, libraries open and contact professionals are allowed to work again.
1st June	Restaurants, bars, museums and theatres are allowed to operate with maximum 30 people.
2nd June	The high schools are allowed to re-operate.
21st June 2020	Public demonstration against lockdown measures held in The Hague.

The Dutch government introduced strict lockdown measures with monetary fines for non-observant parties from 23th March onwards and it continued through April and early May. From 1st June the lockdown measures eased out but masks were mandated while using the public transport. Apparently, the use and observance of lockdown measures and stay home behavior were primary interventions to stop the rapidly surging virus cases in the Netherlands and subsequently the Covid-19 survey also elicited information about the stay home behavior that is discussed in the next section. Another important event is public opposition in the form of demonstrations against the lockdown measures that happened around the survey time in June. This public opposition can play an important role in the willingness to accept future lockdown measures and this issue is discussed in the next section.

2.1. The dependent variables

This section reports the survey questions that formulate the dependent variables in this study. There are three dependent variables that measure staying home behavior under three different scenarios. The first two questions measure the attitude towards the appropriateness of staying home behavior while the last question elicits information about the intended stay home behavior in case the government mandates it. The exact questions are reported below.7

a) Stay home if any symptoms

During the lockdown in April, it was appropriate for people in the Netherlands who had any symptoms of a coronavirus infection (e.g. cold, sore throat, cough, fever) to stay at home and avoid social contacts. To what extent do you agree with statement 1?

• 1 = Strongly Disagree; 5 = Strongly Agree

b) Stay home if no symptoms

During the lockdown in April, it was appropriate for people in the Netherlands who had no symptoms of a coronavirus infection (e.g. cold, sore throat, cough, fever) to stay at home and avoid social contacts. To what extent do you agree with statement 1?

• 1 = Strongly Disagree; 5 = Strongly Agree

c) Stay home if government mandates

If the government would mandate everyone to stay home tomorrow (except for basic grocery shopping), would you stay at home?

• 1 = Yes; 2 = No; 3 = I work in a crucial profession myself

For the first two dependent variables that measure attitude towards the appropriateness of stay home behavior, the distribution of response is highly skewed towards ‘Agree’ and ‘Strongly Agree’ choices (please see Appendix 1). For the first outcome variable (Stay home if any symptoms), almost 89% while for the second outcome variable (Stay home if no symptoms), 70% of the respondents are accounted by only two categories (Agree + Strongly Agree) while the other three categories have smaller number of observations. The existing literature (for example: DiStefano, Shi, & Morgan, 2021; Linacre, 2002; Muis, Winne, & Edwards, 2009) suggests merging the adjacent categories with low observations to remedy the data distribution. Furthermore, Linacre (2002) discusses the possibility that response in case of socially-acceptable propositions can possibly pivot around ‘Agree’ and ‘Strongly Agree’ categories and a possible solution is to merge these adjacent categories into one group. A recent work by DiStefano et al. (2021) also shows that reducing categories in case of small observations for some groups of the Likert scale enhances the accuracy of the estimated parameters. Therefore, I combined these two responses into one category (1 = Agree + Strongly Agree) and the remaining three responses into one category (0 = Strongly Disagree + Disagree + Neither Agree nor Disagree) to form binary variables for staying home measures. For the third dependent variable, the respondents who reported they work in a crucial profession (n = 305) are dropped from the analysis because they cannot stay home due to their profession. Further details about the dependent variables are reported in Table 2 while details about all control variables are reported in Appendix 1.

Table 2.

Data summary and statistics.

Variable	Observations (N)	Mean / Percentage	Min - Max
Dependent Variables
Stay home if any symptoms (1 = Yes; 0 = No)	2287	0.89	0 – 1
Stay home if no symptoms (1 = Yes; 0 = No)	2286	0.70	0 – 1
Stay home if govt. mandates (1 = Yes; 0 = No)	1980	0.80	0 – 1
Primary Independent Variables
Donated in 2020 (1 = Yes; 0 = No)	2289	0.34	0 – 1
Donated in 2019 (1 = Yes; 0 = No)	2246	0.30	0 – 1
Donated in 2015 (1 = Yes; 0 = No)	1771	0.30	0 – 1
Donated in 2010 (1 = Yes; 0 = No)	1222	0.38	0 – 1
Trust in 2020	2242	6.05 (SD = 2.22)	0 – 10
Trust in 2019	2246	5.95 (SD = 2.21)	0 – 10
Trust in 2015	1733	6.03 (SD = 2.22)	0 – 10
Trust in 2010	1207	6.18 (SD = 2.16)	0 – 10

SD = Standard Deviation.

There are several limitations of the outcome variables that are important to discuss here. The first two outcome variables ask respondents to state how appropriate it was for people to stay home during the lockdown in April. These questions do not directly inquire about the own social distancing behavior of the respondents, rather elicit their personal opinions and attitude towards the appropriateness of staying home behavior and can also capture whether they approve of the lockdown. But as personal opinions are likely to be self-exercised as well, the first two questions are more likely to represent the respondent behavior.

The third outcome variable elicits direct information about the willingness or intention to observe lockdown in case the Dutch government imposes it tomorrow. The response to this question is likely to be influenced the contextual factors at the time of the survey, the most important and relevant ones are the Covid-19 severity and phasing out of the lockdown restrictions. The survey was implemented in June 2020 and by that time the Covid-19 severity was decreasing in the Netherlands. Also, as reported in Table 1, most lockdown restrictions were eased out in June and schools, restaurants, bars, museums, libraries and theatres re-opened with social distancing protocols in place. In the latter half of June, there was also public demonstration against the lockdown measures. All these factors can influence the response to the third outcome variable and subsequently influence the relation between staying home behavior and social preferences.

2.3. Primary explanatory variables

The prosocial behavior and trust are primary explanatory variables. They are operationalized as follows.

Prosocial behavior = 1 if the respondent in the last 12 months donated money to at least one organization working for a noble cause and zero otherwise.

The list of donation receiving organizations used in the survey to elicit prosocial behavior contained the following items: 1) sports club or club for outdoor activities 2) cultural association or hobby club 3) trade union 4) business, professional or agrarian organization 5) consumer organization or automobile club 6) humanitarian aid, human rights, minorities/migrants organization 7) environmental protection, peace or animal rights organization 8) religious organization 9) political party 10) academic organization 11) social service organizations (for youth, pensioners, women or friends) or 12) other organizations. The survey did not elicit the amount of money donated to these organizations, rather it gathered information whether a respondent donate or did not donate to any of these organizations. Therefore, the operationalization of prosocial behavior used in this study is also binary in nature. The donation space is not restricted to the typical charity organizations because it can significantly influence the prosocial behavior either in the upward or downward direction. A recent meta-analysis by Umer, Kurosaki, and Iwasaki (2022) also shows that prosocial behavior is quite sensitive to the nature of the recipient. A comparison of proportions of donors donating to all these organizations across the four years (2020, 2019, 2015 and 2010) is provided in Appendix 1A.

The post-Covid-19 prosocial behavior for the last 12 months is elicited in October-November in 2020 and hence overlaps with two months pre-Covid-19. This can be an issue and cause endogeneity because several studies report change in pro-sociality post Covid-19 (for example: Shachat et al., 2021; Brañas-Garza et al., 2022).8 To overcome these issues and as a robustness test, I use prosocial information collected in 2019 (pre-Covid-19) to examine whether prosocial behavior influences staying home behavior.

Trust= Generally speaking, would you say that most people can be trusted, or that you can't be too careful in dealing with people? Please indicate a score of 0 to 10.

• 0 = You can't be too careful

• 10 = Most people can be trusted

• -9 = I don't know

Further details about prosocial behavior and trust measured in 2020, 2019, 2015 and 2010 are reported in Table 2 (Primary Independent Variables). 47 respondents did not provide answer for the trust question (-9 = I don't know) in 2020, 38 in 2015 while 15 did not provide answer in 2010 and they are removed from the analysis.9

2.4. Statistical analysis

The primary objective is to analyze whether donor proportion (pro-sociality) and general trust predicts staying home behavior during lockdown. To do so, OLS regressions of the following type are estimated:

$$Y_i={\beta }_0+{\beta }_{1}Donate{d}_i+{\beta }_i{X}_i+{ϵ}_i$$ (1)

$$Y_i={\beta }_0+{\beta }_{2}Trus{t}_i+{\beta }_i{X}_i+{ϵ}_i$$ (2)

Where $Y_i$ in Eq. (1) and 2 is the dependent variable and corresponds to one of the three outcome variables (stay home if any symptoms; stay home if no symptoms; stay home if government mandates) measured in 2020. The variable $Donate{d}_i$ in Eq. (1) is a dummy variable that takes a value of 1 if the respondent i donated in last 12 months and zero otherwise. The variable $Trus{t}_i$ represents the general trust level of individual i. Both Donated and Trust measured in 2020, 2019, 2015 and 2010 are used in separate regressions to analyze their effect on stay home behavior in 2020. A positive and significant ${\beta }_1$ and ${\beta }_2$would mean pro-sociality and trust positively predict staying home behavior. The vector $X_i$ is a set of controls observed in 2020 that include gender, age, number of children, whether respondent lives with partner, marital status, education, housing type, primary occupation and gross monthly income category10 while ${ϵ}_i$ represents the random error term. Robust standard errors clustered around individual IDs are used in the OLS regressions. The analysis is performed using STATA 16.

3. Results

3.1. Pro-sociality and staying home behavior

The results for pro-sociality as primary explanatory variable are in Table 3 while complete results with all controls are in Appendices 3 – 6. The prosocial behavior measured in 2020 has a significant and positive effect on staying home attitude in case of virus symptoms and in case of no virus symptoms (Table 3: Panel A, Regression 1 & 3). The effect however becomes insignificant once controls are added to the regression (Panel A, Regression 2 & 4). As discussed earlier, pro-sociality itself can be influenced by the pandemic. Therefore, I also examined the effect of pre-pandemic pro-sociality measured in 2019. The pro-sociality in 2019 positively and significantly predicts attitude towards staying home in case of virus symptoms (Panel B, Regressions 1), and the effect remains marginally significant after the addition of controls to the regression (Panel B, Regression 2).

Table 3.

Prosocial behavior & covid-19 preventive measures.

	Stay home if any symptoms	Stay home if any symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]
Panel A: Prosocial Behavior in 2020 & Preventive Behavior
Donated in 2020	0.037***	0.018	0.045**	0.025	0.012	−0.003
	(0.013)	(0.013)	(0.020)	(0.020)	(0.019)	(0.019)
Controls	No	Yes	No	Yes	No	Yes
Observations	2287	2287	2286	2286	1980	1980
R-squared	0.003	0.041	0.002	0.033	0.000	0.041
Panel B: Prosocial Behavior in 2019 & Preventive Behavior
Donated in 2019	0.039***	0.025*	−0.005	−0.021	0.021	0.014
	(0.014)	(0.014)	(0.021)	(0.021)	(0.020)	(0.020)
Controls	No	Yes	No	Yes	No	Yes
Observations	2244	2244	2243	2243	1942	1942
R-squared	0.003	0.042	0.000	0.033	0.001	0.043
Panel C: Prosocial Behavior in 2015 & Preventive Behavior
Donated in 2015	0.040***	0.027*	0.028	0.020	0.013	0.005
	(0.015)	(0.015)	(0.023)	(0.024)	(0.022)	(0.023)
Controls	No	Yes	No	Yes	No	Yes
Observations	1769	1769	1769	1769	1533	1533
R-squared	0.004	0.030	0.001	0.031	0.000	0.048
Panel D: Prosocial Behavior in 2010 & Preventive Behavior
Donated in 2010	0.020	0.005	0.014	−0.006	0.004	−0.013
	(0.017)	(0.019)	(0.026)	(0.027)	(0.025)	(0.026)
Controls	No	Yes	No	Yes	No	Yes
Observations	1221	1221	1221	1221	1077	1077
R-squared	0.001	0.036	0.000	0.040	0.000	0.063

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Overall, there is very weak evidence that pro-sociality measured in the short-run predicts attitude towards the appropriateness of staying home behavior in the Netherlands. This result does not support the earlier findings by Campos-Mercade et al. (2021) who find strong evidence that pro-sociality measured pre and post-Covid-19 strongly predicts staying home behavior. Campos-Mercade et al. (2021) examine behavior in Sweden while the current study uses data from the Netherlands. People in Sweden generally observed preventive measures following government requests while the situation is completely different in the Netherlands where people have protested against the lockdown measures. This contextual difference which ultimately reflects in the health behaviors of people in the two countries can be a key element disproportionately affecting the relation between pro-sociality and staying home behavior in the two European countries. Furthermore, the instruments used to measure the preventive behaviors and pro-sociality in the current study are not fully comparable to the work of Campos-Mercade et al. (2021) and this can also be a possible reason for divergent results between the two studies. Nevertheless, current findings based on a systematic and well-structured data collection make it clear that pro-sociality measured in the short-run does not robustly predict attitude towards staying home in the Netherlands. These findings also reflect that the effectiveness of prosocial behavior in predicting staying home behavior is heterogeneous across countries even within Europe.

Panel C and D in Table 3 report the outcomes for the long-run relationship between pro-sociality and staying home behavior. The pro-sociality measured in 2015 positively predicts the attitude towards the appropriateness of staying home behavior in case of virus symptoms in both regressions without and with controls (Panel C, Regression 1 & 2). On the other hand, there is no significant evidence that pro-sociality measured in 2010 predicts attitudes or intentions towards staying home behavior. Overall, these results provide marginal evidence for the long-run role of pro-sociality in predicting the stay home attitude and intentions in the Netherlands.

3.2. Trust and staying home behavior

The findings for the effect of trust on staying home behavior are summarized in Table 4 while complete results are provided in Appendices 7 – 10. For the short-term analysis, trust measured in 2020 positively predicts the attitude towards the appropriateness of staying home in case of symptoms and in case of no symptoms, and the effect remains significant even when controls are added to the regression (Panel A: Regressions 1–4). Trust measured in 2019 also positively and significantly predicts the attitude towards the appropriateness of staying home in case of symptoms and in case of no symptoms in regressions without and with controls (Panel B: Regressions 1–4). It also positively and significantly predicts the intended staying home in case the government mandates it, however the effect becomes insignificant once controls are added. Overall, whether we use trust information from 2020 or 2019, they both predict the attitude towards staying home behavior in case of symptoms and in case of no symptoms. In comparison to the results for pro-social behavior discussed in the previous section, trust is relatively a stable predictor of staying home behavior in the short-run.

Table 4.

Trust & Covid-19 preventive measures.

	Stay home if any symptoms	Stay home if any symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]
Panel A: Effect of Trust in 2020 on Preventive Behavior
Trust in 2020	0.011***	0.007**	0.013***	0.009*	0.005	0.000
	(0.003)	(0.003)	(0.004)	(0.005)	(0.004)	(0.005)
Controls	No	Yes	No	Yes	No	Yes
Observations	2240	2240	2239	2239	1936	1936
R-squared	0.007	0.042	0.004	0.030	0.001	0.040
Panel B: Effect of Trust in 2019 on Preventive Behavior
Trust in 2019	0.011***	0.007**	0.014***	0.009*	0.007*	0.004
	(0.003)	(0.003)	(0.004)	(0.005)	(0.004)	(0.004)
Controls	No	Yes	No	Yes	No	Yes
Observations	2244	2244	2243	2243	1942	1942
R-squared	0.006	0.043	0.005	0.034	0.002	0.043
Panel C: Effect of Trust in 2015 on Preventive Behavior
Trust in 2015	0.012***	0.008**	0.014***	0.012**	0.002	0.001
	(0.004)	(0.004)	(0.005)	(0.005)	(0.005)	(0.005)
Controls	No	Yes	No	Yes	No	Yes
Observations	1731	1731	1731	1731	1504	1504
R-squared	0.008	0.034	0.005	0.034	0.000	0.049
Panel D: Effect of Trust in 2010 on Preventive Behavior
Trust in 2010	0.013***	0.010**	0.018***	0.016***	0.006	0.002
	(0.004)	(0.004)	(0.006)	(0.006)	(0.006)	(0.006)
Controls	No	Yes	No	Yes	No	Yes
Observations	1206	1206	1206	1206	1064	1064
R-squared	0.009	0.041	0.007	0.046	0.001	0.062

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

For the long-term analysis, trust measured in 2015 and 2010 positively predicts the attitude towards the appropriateness of staying home behavior in case of symptoms as well as in case of no symptoms and the coefficient is significant in regressions without and with controls (Panel C & D: Regressions 1–4). Overall, the long-term analysis leads to almost similar results as those for the short-term analysis and reveals that trust is a stable predictor for voluntary staying home behavior in the long-run as well.11

While there is evidence that pro-sociality weakly and trust strongly predicts the attitude towards the appropriateness of staying home behavior, both pro-sociality and trust do not have significant association with the intended staying home behavior in case the government mandates a lockdown. The insignificant association between preferences and intended observance of stay home orders from the government is likely to be driven by the timing of the survey which coincided with the easing out of lockdown restrictions and a decrease in the Covid-19 severity in the Netherlands. Eliciting willingness to observe government-imposed lockdown at a time when the socioeconomic activities resumed in the Netherlands can lead to a possible downward bias in response and undermine the important role of social preferences in predicting the willingness to follow the government-imposed lockdowns. The findings however sink in with the public demonstrations in some Dutch cities against the lockdown measures at the time of survey (reported in Table 1) as well as the current situation in the Netherlands where a significant opposition from all across the country against the government-imposed lockdowns was observed primarily in the form of protests at the start of this year.

3.3. Robustness check

The number of matched respondents decrease as we move from 2020 to 2010. This is primarily because some of the respondents present in 2010 did not participate in the subsequent surveys. While the gap between the number of respondents in 2020 and 2019 is marginal, the difference between respondents in 2020 and 2015 as well as 2010 respondents is quite large. Apart from the time difference, different sample sizes in the short and long-run can also influence results. Therefore, I re-perform analysis for all years using identical number of observations to examine the robustness of outcomes.

The results for pro-sociality and preventive behavior are summarized in Table 5 while complete results are available in Appendices 13-15. The regressions with identical observations for all four years lead to very similar outcomes as reported in Table 3: pro-sociality in 2019 and 2015 still positively and significantly predict attitudes towards the appropriateness of stay home behavior in case of virus symptoms in regressions without and with controls (Table 5, Panel A: Regressions 3–6).

Table 5.

Prosocial behavior & Covid-19 preventive measures (identical estimates).

Regression#	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]
Panel A: Stay home if any symptoms
Donated in 2010	0.02	0.005
	(0.017)	(0.019)
Donated in 2015			0.040**	0.031*
			(0.017)	(0.018)
Donated in 2019					0.045***	0.036**
					(0.017)	(0.018)
Donated in 2020							0.038**	0.019
							(0.017)	(0.018)
Controls	No	Yes	No	Yes	No	Yes	No	Yes
Observations	1221	1221	1221	1221	1221	1221	1221	1221
R-squared	0.001	0.036	0.004	0.038	0.005	0.038	0.004	0.036
Panel B: Stay home if no symptoms
Donated in 2010	0.014	(0.006)
	(0.026)	(0.027)
Donated in 2015			0.023	0.018
			(0.028)	(0.029)
Donated in 2019					0.015	0.011
					(0.028)	(0.029)
Donated in 2020							0.046*	0.029
							(0.027)	(0.028)
Controls	No	Yes	No	Yes	No	Yes	No	Yes
Observations	1221	1221	1221	1221	1221	1221	1221	1221
R-squared	0.000	0.04	0.001	0.04	0.000	0.04	0.002	0.04
Panel C: Stay home if govt. mandates
Donated in 2010	0.004	(0.013)
	(0.025)	(0.026)
Donated in 2015			(0.003)	(0.013)
			(0.027)	(0.027)
Donated in 2019					0.024	0.018
					(0.027)	(0.027)
Donated in 2020							0.022	0.013
							(0.025)	(0.026)
Controls	No	Yes	No	Yes	No	Yes	No	Yes
Observations	1077	1077	1077	1077	1077	1077	1077	1077
R-squared	0.000	0.063	0.000	0.063	0.001	0.063	0.001	0.063

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

The results for trust and preventive behavior are summarized in Table 6 while complete results with all controls are reported in Appendices 16-18. The results for stay home behavior in case of symptoms support the earlier findings reported in Table 4: trust in all four years has a positive and significant influence on attitudes towards the appropriateness of stay home behavior in regressions without and with controls. Similar results are observed for stay home behavior in case of no virus symptoms for 2010, 2015 and 2019. Only for the year 2020, the coefficient is insignificant (Table 6, Panel B: Regressions 7–8). Overall, the results for trust as a predictor of stay home behavior with identical sample are quite similar to the ones reported in Table 4. The robustness checks for both pro-sociality and trust support the earlier findings and reinforce that trust in comparison to pro-sociality is a stable predictor of attitudes towards the appropriateness of stay home behavior in the Netherlands.

Table 6.

Trust & Covid-19 preventive measures (identical estimates).

Regression#	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]
Panel A: Stay home if any symptoms
Trust in 2010	0.012***	0.009**
	(0.004)	(0.004)
Trust in 2015			0.013***	0.010**
			(0.004)	(0.004)
Trust in 2019					0.013***	0.010**
					(0.004)	(0.004)
Trust in 2020							0.012***	0.010**
							(0.004)	(0.004)
Controls	No	Yes	No	Yes	No	Yes	No	Yes
Observations	1152	1152	1152	1152	1152	1152	1152	1152
R-squared	0.008	0.041	0.010	0.043	0.010	0.042	0.009	0.042
Panel B: Stay home if no symptoms
Trust in 2010	0.018***	0.017***
	(0.006)	(0.006)
Trust in 2015			0.014**	0.013**
			(0.006)	(0.006)
Trust in 2019					0.016***	0.013**
					(0.006)	(0.007)
Trust in 2020							0.009	0.006
							(0.006)	(0.006)
Controls	No	Yes	No	Yes	No	Yes	No	Yes
Observations	1152	1152	1152	1152	1152	1152	1152	1152
R-squared	0.008	0.043	0.005	0.040	0.006	0.040	0.002	0.037
Panel C: Stay home if govt. mandates
Trust in 2010	0.008	0.003
	(0.006)	(0.006)
Trust in 2015			0.007	0.005
			(0.006)	(0.006)
Trust in 2019					0.011*	0.007
					(0.006)	(0.006)
Trust in 2020							0.008	0.004
							(0.006)	(0.006)
Controls	No	Yes	No	Yes	No	Yes	No	Yes
Observations	1018	1018	1018	1018	1018	1018	1018	1018
R-squared	0.002	0.06	0.001	0.061	0.003	0.061	0.002	0.060

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

4. Conclusions and discussions

The success of fight against the Covid-19 and the economic recovery is largely dependent on the socially responsible behavior and voluntary compliance of citizens. This study examined the short and long-term role of preferences (pro-sociality and trust) in predicting attitudes towards the appropriateness of staying home behavior in the presence and absence of virus symptoms and the intended staying home behavior if the government mandates it in the Netherlands. The main findings are that both pro-sociality and trust predict attitudes towards the appropriateness of staying home behavior. However, the relative comparison shows that trust in comparison to pro-sociality in both short and long-run is a better and stable predictor of attitude towards staying home behavior. The results for the pro-social behavior do not completely sink in with those from Sweden reported by Campos-Mercade et al. (2021) and suggest that the behavior is quite heterogeneous even within Europe. Therefore, lessons learned from a specific European country need be scrutinized before generalizing them for other countries even within Europe. The findings also underscore the relative applicability of trust in the health economics domain.

The relatively stable performance of trust in both short and long-run in predicting attitudes towards staying home behavior leads to important implications for health crisis management in both short and long run specifically in the Dutch context. First, relying on the geographic differences, the government can identify regions where people have high levels of general trust and effectively monitor staying home behavior with subtle nudges rather than relying on force in those areas. This sub-group specific policy can save monetary cost and prevent social unrest specifically in the high trust regions. Second, based on the regional trust levels, the government can identify high risk areas where people are less likely to voluntarily comply to the staying home orders. More attention both in terms of resources and enforcement tools can be directed towards these regions to ensure a uniform observance of preventive measures across the Netherlands. These high-risk regions also offer an opportunity to the policy makers to further examine why trust in these regions is lower and how it can be improved in the future.

On the academic front, the current findings reveal the importance and applicability of social preferences in predicting health behaviors. Therefore, future health research can further enrich the existing models used to examine disease and virus spread by including trust in them. Also, trust and pro-sociality might be effectively used to predict behavior in other health domains such as smoking that have externalities, and further work in this direction can lead to interesting outcomes. Lastly, while it is difficult to generalize current results to other countries, it will be interesting to examine in the future research whether trust in comparison to pro-sociality is a better predictor of preventive measures at a global scale.

Data statement

This paper we make use of data of the LISS (Longitudinal Internet studies for the Social Sciences) panel administered by CentERdata (Tilburg University, The Netherlands).

The LISS panel data were collected by CentERdata (Tilburg University, The Netherlands) through its MESS project funded by the Netherlands Organization for Scientific Research.

For data access, please visit the LISS website at https://www.lissdata.nl

Analysis code

The STATA code is available in the online supplemental materials.

Appendix 1. Summary & Stats of Variables Observed in 2020.

Variables	Observations	Mean / Percentage
Dependent Variables
Stay home if any symptoms	2287	4.40	1 – 5
1 = Strongly disagree	71	3.10%
2 = Disagree	49	2.14%
3 = Neither agree nor disagree	135	5.90%
4 = Agree	662	28.95%
5 = Strongly agree	1370	59.90%
Stay home if no symptoms	2286	3.78	1 – 5
1 = Strongly disagree	81	3.54%
2 = Disagree	240	10.50%
3 = Neither agree nor disagree	363	15.88%
4 = Agree	1009	44.14%
5 = Strongly agree	593	25.94%
Controls
Male	1108	48.41%
Age (Years)	2289	56.82 (SD = 17.38)	17 – 103
Children	2289	0.61 (SD = 1.02)	0 – 6
Number of HH members	2289	2.38 (SD = 1.22)	1 – 8
HH head lives with partner (1 = Yes; 0 = No)	2289	0.703	0 – 1
Marital Status
Married	1297	56.66%
Separated	7	0.31%
Divorced	254	11.10%
Widow/ Widower	145	6.33%
Never Married	586	25.60%
Education
Primary School	63	2.75%
Junior & Senior High School	632	27.61%
Junior College & College	1195	52.21%
University	366	15.99%
Other	25	1.09%
Not yet started education	8	0.35%
Housing Type
Self-owned	1649	72.04%
Rental	620	27.09%
Cost-free	16	0.70%
Unknown	4	0.17%
Primary Occupation
Paid Work (Employed + family business + freelancer)	1011	44.17%
Unemployed (Job-seeker + non-seekers)	58	2.53%
Student	115	5.02%
Unpaid Work (housekeeping + voluntary work)	244	10.66%
Pensioner	757	33.07%
Work Disability	94	4.11%
Does something else	10	0.44%
Gross Monthly Income (Euros)
No Income	152	6.64%
500 or less	27	1.18%
501 to 1000	81	3.54%
1001 to 1500	112	4.89%
1501 to 2000	155	6.77%
2001 to 2500	149	6.51%
2501 to 3000	143	6.25%
3001 to 3500	144	6.29%
3501 to 4000	101	4.41%
4001 to 4500	66	2.88%
4501 to 5000	56	2.45%
5001 to 7500	99	4.33%
More than 7500	22	0.96%
Don't know	982	42.90%

SD = Standard Deviation. The variable ‘Education’ is recoded reduce the number of categories such that junior high school and senior high school are grouped together while junior college and college are grouped together. Similarly, the variable ‘Income’ is recoded to reduce the number of categories. The income categories recoded are ‘Paid Work’ that includes employed people, involved in business or freelancing. ‘Unemployed’ includes those looking for job as well as those not seeking a job while ‘Unpaid Work’ includes respondents doing housekeeping or voluntary work.

Appendix 1A. Temporal Comparison of Proportion of People Donating to Different Organizations.

	Donor Proportions (Observations per year = 1220)
Variable	2020 versus 2019	2020 versus 2015	2020 versus 2010
Sports club or club for outdoor activities	2% vs 1.5%	2% vs 2%	2% vs 2.7%
Cultural association or hobby club	2.6% vs 2.6%	2.6% vs 1.6% *	2.6% vs 3.1%
Trade union	0.1% vs 1.2% ***	0.1% vs 0.7% **	0.1% vs 0.7% **
Business, professional or agrarian organization	0.2% vs 0.4%	0.2% vs 0.6% *	0.2% vs 0.7% **
Consumer organization or automobile club	1.2% vs 0.8%	1.2% vs 0.7%	1.2% vs 0.7%
Humanitarian aid, human rights, minorities/ migrant organization	22.3% vs 18.7% **	22.3% vs 17.4% ***	22.3% vs 26.1% **
Environmental protection, peace or animal rights organization	15.7% vs 14.3%	15.7% vs 14.3%	15.7% vs 17.4%
Religious organization	10.7% vs 7.5% ***	10.7% vs 8.2% **	10.7% vs 12%
Political party	1.6% vs 1.3%	1.6% vs 1.0%	1.6% vs 1.2%
academic organization	1.2% vs 1.6%	1.2% vs 1.4%	1.2% vs 1.5%
Social service organizations (for youth, pensioners, women or friends)	1.2% vs 1.2%	1.2% vs 0.7%	1.2% vs 1%
Other organizations	2.8% vs 2.2%	2.8% vs 1.9%	2.8% vs 3.1%

• Note: The survey for 2020 decomposes donations to humanitarian aid, human rights, migrants into two categories: Humanitarian aid or human rights organization and Migrants. The other three survey waves treat these two categories as one group. Therefore, these two categories for 2020 survey are combined together to make the survey item comparable to other three waves (2019, 2015 and 2010).

• The percentage in each cell represents the proportion of people who donated to that organization in the two years under consideration. The between year comparison is performed with the help of two-sided test of proportions such that proportion in 2019, 2015 and 2010 are separately compared with proportion in 2020. Only those respondents who participated in all four survey waves are considered for the comparisons. This restriction is applied to eliminate the possible effect of sample size differences on the comparison of proportions.

• Significant differences in proportions are indicated by asterisk. *** Significant at 1%, ** Significant at 5%, * Significant at 10%.

Appendix 2. Examining differences between respondents who answer and those do not answer the trust question.

Variables	Respondent answered the trust question		Respondent did not answer the trust question
	Mean (SD)	N	Mean (SD)	N	t-stat (p-value)
Year = 2020
Gender	1.51 (0.50)	2242	1.53 (0.50)	47	−0.22 (0.82)
Age	56.79 (17.34)	2242	58.15 (19.03)	47	−0.53 (0.60)
Children	0.61 (1.02)	2242	0.62 (1.11)	47	−0.02 (0.98)
Family Size	2.38 (1.22)	2242	2.38 (1.29)	47	−0.04 (0.97)
Marital Status	2.44 (1.75)	2242	2.19 (1.69)	47	0.98 (0.33)
Education	2.86 (0.78)	2242	2.72 (0.83)	47	1.23 (0.22)
Year = 2015
Gender	1.50 (0.50)	1733	1.53 (0.51)	38	−0.30 (0.77)
Age	59.55 (15.75)	1733	59.13 (19.02)	38	0.16 (0.87)
Children	0.53 (0.95)	1733	0.79 (1.19)	38	−1.67 (0.10) *
Family Size	2.29 (1.12)	1733	2.58 (1.37)	38	−1.54 (0.12)
Marital Status	2.23 (1.66)	1733	2.45 (1.74)	38	−0.79 (0.43)
Education	2.84 (0.76)	1733	2.87 (0.88)	38	−0.21 (0.84)
Year = 2010
Gender	1.48 (0.50)	1207	1.67 (0.49)	15	−1.46 (0.14)
Age	62.87 (13.60)	1207	60.2 (15.76)	15	0.76 (0.45)
Children	0.46 (0.90)	1207	0.73 (1.44)	15	−1.15 (0.25)
Family Size	2.25 (1.05)	1207	2.27 (1.75)	15	−0.05 (0.96)
Marital Status	2.01 (1.54)	1207	2.93 (1.94)	15	−2.29 (0.02) **
Education	2.77 (0.75)	1207	2.93 (1.22)	15	−0.81 (0.42)

SD = Standard Deviation. T-stat is from test of means. ** p < 0.05, * p < 0.10.

Appendix 3. Prosocial Behavior in 2020 & Preventive Behavior.

	Stay home if any symptoms	Stay home if any symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]
Donated in 2020	0.037***	0.018	0.045**	0.025	0.012	−0.003
	(0.013)	(0.013)	(0.020)	(0.020)	(0.019)	(0.019)
Male		−0.042***		−0.044**		−0.075***
		(0.015)		(0.021)		(0.021)
Age (Years)		0.002**		0.002**		0.001
		(0.001)		(0.001)		(0.001)
Children		−0.047**		−0.038		−0.018
		(0.020)		(0.030)		(0.027)
Family Size		0.024		0.012		0.007
		(0.018)		(0.028)		(0.025)
Lives with Partner (Base = Yes)		−0.014		0.012		−0.003
		(0.030)		(0.042)		(0.040)
Marital Status (Base: Married)
Separated		0.122***		−0.182		−0.249
		(0.037)		(0.193)		(0.198)
Divorced		0.042		−0.024		−0.046
		(0.027)		(0.042)		(0.041)
Widow or widower		−0.038		0.024		−0.023
		(0.038)		(0.047)		(0.044)
Never Married		0.028		−0.007		−0.066*
		(0.022)		(0.034)		(0.035)
Education (Base: Junior College & College)
Primary School		0.009		−0.182*		−0.119
		(0.103)		(0.105)		(0.095)
Junior & Senior High School		0.130		−0.054		−0.097
		(0.089)		(0.087)		(0.080)
University		0.156*		−0.032		−0.095
		(0.089)		(0.086)		(0.080)
Other		0.194**		0.006		−0.069
		(0.089)		(0.089)		(0.083)
Not started		0.055		0.170		−0.007
		(0.171)		(0.149)		(0.165)
Housing Type (Base: Self-Owned)
Rental		−0.001		−0.012		−0.061***
		(0.017)		(0.024)		(0.023)
Cost Free		0.049		0.004		−0.049
		(0.061)		(0.118)		(0.125)
Unknown		0.133***		−0.161		0.216***
		(0.033)		(0.263)		(0.047)
Occupation (Base: Paid Work)
Unemployed		0.007		−0.055		−0.051
		(0.044)		(0.068)		(0.067)
Student		0.017		0.012		0.098*
		(0.046)		(0.060)		(0.059)
Unpaid Work		0.002		0.059		0.068*
		(0.030)		(0.040)		(0.038)
Pensioner		−0.004		0.038		0.047
		(0.022)		(0.033)		(0.033)
Work Disability		−0.017		−0.007		0.042
		(0.038)		(0.053)		(0.047)
Something else		−0.162		−0.148		−0.105
		(0.147)		(0.153)		(0.169)
Income in Euros (Base: 2001 - 2500 Euros)
No Income		−0.040		−0.059		−0.087
		(0.045)		(0.060)		(0.058)
500 or less		−0.001		0.005		0.042
		(0.070)		(0.100)		(0.084)
501 to 1000		0.021		0.036		−0.008
		(0.038)		(0.060)		(0.056)
1001 to 1500		0.007		−0.047		0.004
		(0.036)		(0.057)		(0.054)
1501 to 2000		−0.021		−0.019		−0.023
		(0.035)		(0.053)		(0.051)
2501 to 3000		0.011		−0.016		−0.049
		(0.034)		(0.054)		(0.055)
3001 to 3500		−0.011		−0.029		−0.040
		(0.035)		(0.054)		(0.055)
3501 to 4000		0.043		−0.002		−0.007
		(0.034)		(0.060)		(0.059)
4001 to 4500		0.005		0.023		0.139***
		(0.043)		(0.066)		(0.053)
4501 to 5000		0.012		−0.001		0.010
		(0.046)		(0.071)		(0.069)
5001 to 7500		0.021		−0.063		−0.011
		(0.036)		(0.063)		(0.059)
More than 7500		−0.001		−0.047		−0.008
		(0.065)		(0.106)		(0.100)
Don't Know		−0.025		−0.038		−0.025
		(0.026)		(0.041)		(0.039)
Constant	0.876***	0.623***	0.685***	0.632***	0.791***	0.905***
	(0.009)	(0.113)	(0.012)	(0.127)	(0.011)	(0.122)
Controls	No	Yes	No	Yes	No	Yes
Observations	2287	2287	2286	2286	1980	1980
R-squared	0.003	0.041	0.002	0.033	0.000	0.041

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 4. Prosocial Behavior in 2019 & Preventive Behavior.

	Stay home if any symptoms	Stay home if any symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]
Donated in 2019	0.039***	0.025*	−0.005	−0.021	0.021	0.014
	(0.014)	(0.014)	(0.021)	(0.021)	(0.020)	(0.020)
Male		−0.041***		−0.045**		−0.069***
		(0.015)		(0.021)		(0.021)
Age (Years)		0.002**		0.002**		0.001
		(0.001)		(0.001)		(0.001)
Children		−0.046**		−0.034		−0.019
		(0.020)		(0.030)		(0.027)
Family Size		0.023		0.012		0.007
		(0.018)		(0.028)		(0.025)
Lives with Partner (Base = Yes)		−0.007		0.021		0.003
		(0.030)		(0.042)		(0.041)
Marital Status (Base: Married)
Separated		0.115***		−0.191		−0.253
		(0.038)		(0.195)		(0.196)
Divorced		0.034		−0.031		−0.053
		(0.027)		(0.042)		(0.042)
Widow or widower		−0.045		0.022		−0.029
		(0.038)		(0.047)		(0.044)
Never Married		0.023		−0.001		−0.070*
		(0.023)		(0.035)		(0.036)
Education (Base: Junior College & College)
Primary School		0.003		−0.188*		−0.132
		(0.105)		(0.106)		(0.095)
Junior & Senior High School		0.132		−0.055		−0.101
		(0.090)		(0.087)		(0.080)
University		0.159*		−0.031		−0.098
		(0.089)		(0.086)		(0.079)
Other		0.198**		0.014		−0.074
		(0.090)		(0.089)		(0.082)
Not started		0.060		0.164		−0.005
		(0.172)		(0.151)		(0.166)
Housing Type (Base: Self-Owned)
Rental		−0.003		−0.013		−0.062***
		(0.017)		(0.024)		(0.024)
Cost Free		0.051		0.008		−0.048
		(0.061)		(0.116)		(0.125)
Unknown		0.133***		−0.160		0.216***
		(0.034)		(0.262)		(0.046)
Occupation (Base: Paid Work)
Unemployed		0.003		−0.033		−0.064
		(0.045)		(0.068)		(0.068)
Student		0.014		−0.004		0.092
		(0.047)		(0.060)		(0.060)
Unpaid Work		0.003		0.062		0.073*
		(0.030)		(0.040)		(0.038)
Pensioner		−0.003		0.038		0.050
		(0.022)		(0.034)		(0.033)
Work Disability		−0.022		0.006		0.040
		(0.039)		(0.053)		(0.048)
Something else		−0.191		−0.214		−0.140
		(0.160)		(0.160)		(0.183)
Income in Euros (Base: 2001 - 2500 Euros)
No Income		−0.037		−0.063		−0.097*
		(0.046)		(0.060)		(0.058)
500 or less		0.008		0.014		0.037
		(0.069)		(0.100)		(0.084)
501 to 1000		0.018		0.031		−0.017
		(0.039)		(0.062)		(0.057)
1001 to 1500		0.013		−0.052		−0.003
		(0.036)		(0.058)		(0.054)
1501 to 2000		−0.020		−0.025		−0.033
		(0.036)		(0.053)		(0.051)
2501 to 3000		0.008		−0.024		−0.062
		(0.035)		(0.055)		(0.056)
3001 to 3500		−0.005		−0.020		−0.064
		(0.035)		(0.055)		(0.056)
3501 to 4000		0.043		−0.003		−0.018
		(0.034)		(0.060)		(0.059)
4001 to 4500		0.004		0.015		0.123**
		(0.044)		(0.067)		(0.054)
4501 to 5000		0.014		0.000		−0.002
		(0.046)		(0.072)		(0.069)
5001 to 7500		0.019		−0.064		−0.028
		(0.037)		(0.063)		(0.059)
More than 7500		−0.001		−0.038		−0.023
		(0.065)		(0.104)		(0.100)
Don't Know		−0.022		−0.038		−0.038
		(0.026)		(0.041)		(0.039)
Constant	0.876***	0.615***	0.702***	0.627***	0.787***	0.916***
	(0.008)	(0.114)	(0.011)	(0.128)	(0.011)	(0.123)
Controls	No	Yes	No	Yes	No	Yes
Observations	2244	2244	2243	2243	1942	1942
R-squared	0.003	0.042	0.000	0.033	0.001	0.043

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 5. Prosocial Behavior in 2015 & Preventive Behavior.

	Stay home if any symptoms	Stay home if any symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]
Donated in 2015	0.040***	0.027*	0.028	0.020	0.013	0.005
	(0.015)	(0.015)	(0.023)	(0.024)	(0.022)	(0.023)
Male		−0.030*		−0.030		−0.066***
		(0.016)		(0.024)		(0.024)
Age (Years)		0.002*		0.001		−0.000
		(0.001)		(0.001)		(0.001)
Children		−0.031		0.008		−0.014
		(0.036)		(0.051)		(0.044)
Family Size		0.016		−0.025		0.008
		(0.035)		(0.050)		(0.042)
Lives with Partner (Base = Yes)		0.005		0.018		0.022
		(0.044)		(0.061)		(0.057)
Marital Status (Base: Married)
Separated		0.070*		−0.139		−0.200
		(0.040)		(0.199)		(0.212)
Divorced		0.008		−0.076		−0.051
		(0.031)		(0.049)		(0.048)
Widow or widower		−0.048		0.003		−0.058
		(0.045)		(0.057)		(0.053)
Never Married		0.014		−0.021		−0.100**
		(0.025)		(0.042)		(0.043)
Education (Base: Junior College & College)
Primary School		−0.133		−0.371***		−0.187*
		(0.116)		(0.122)		(0.105)
Junior & Senior High School		0.018		−0.176*		−0.152*
		(0.097)		(0.100)		(0.088)
University		0.038		−0.161		−0.166*
		(0.096)		(0.099)		(0.088)
Other		0.071		−0.135		−0.156*
		(0.098)		(0.103)		(0.092)
Not started		−0.009		−0.039		0.075
		(0.184)		(0.189)		(0.093)
Housing Type (Base: Self-Owned)
Rental		0.014		0.004		−0.055**
		(0.019)		(0.027)		(0.027)
Cost Free		0.022		0.051		−0.054
		(0.085)		(0.139)		(0.159)
Unknown		0.103***		−0.367		0.209***
		(0.039)		(0.283)		(0.071)
Occupation (Base: Paid Work)
Unemployed		0.024		−0.070		−0.059
		(0.049)		(0.080)		(0.080)
Student		0.041		−0.013		0.144
		(0.061)		(0.099)		(0.091)
Unpaid Work		0.006		0.099**		0.126***
		(0.033)		(0.044)		(0.043)
Pensioner		0.001		0.059		0.080**
		(0.024)		(0.038)		(0.038)
Work Disability		−0.028		−0.029		0.060
		(0.043)		(0.059)		(0.053)
Something else		−0.103		−0.081		−0.295
		(0.183)		(0.231)		(0.254)
Income in Euros (Base: 2001 - 2500 Euros)
No Income		−0.007		0.007		−0.080
		(0.050)		(0.070)		(0.065)
500 or less		−0.005		0.069		0.018
		(0.094)		(0.121)		(0.105)
501 to 1000		0.022		0.029		−0.035
		(0.044)		(0.070)		(0.063)
1001 to 1500		0.008		−0.010		−0.041
		(0.040)		(0.064)		(0.062)
1501 to 2000		−0.018		−0.024		−0.042
		(0.040)		(0.060)		(0.057)
2501 to 3000		0.011		0.013		−0.075
		(0.039)		(0.060)		(0.063)
3001 to 3500		0.015		0.031		−0.026
		(0.037)		(0.061)		(0.061)
3501 to 4000		0.068**		0.013		−0.014
		(0.034)		(0.067)		(0.065)
4001 to 4500		0.048		0.053		0.148***
		(0.041)		(0.074)		(0.057)
4501 to 5000		0.010		0.042		−0.014
		(0.053)		(0.078)		(0.077)
5001 to 7500		0.033		−0.061		−0.036
		(0.040)		(0.073)		(0.068)
More than 7500		−0.023		−0.029		0.026
		(0.078)		(0.118)		(0.104)
Don't Know		−0.020		−0.012		−0.045
		(0.029)		(0.046)		(0.043)
Constant	0.885***	0.753***	0.698***	0.825***	0.792***	1.012***
	(0.009)	(0.136)	(0.013)	(0.162)	(0.012)	(0.148)
Controls	No	Yes	No	Yes	No	Yes
Observations	1769	1769	1769	1769	1533	1533
R-squared	0.004	0.030	0.001	0.031	0.000	0.048

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 6. Prosocial Behavior in 2010 & Preventive Behavior.

	Stay home if any symptoms	Stay home if any symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]
Donated in 2010	0.020	0.005	0.014	−0.006	0.004	−0.013
	(0.017)	(0.019)	(0.026)	(0.027)	(0.025)	(0.026)
Male		−0.032		−0.026		−0.077***
		(0.020)		(0.030)		(0.029)
Age (Years)		0.002		0.001		−0.000
		(0.001)		(0.002)		(0.002)
Children		−0.029		−0.037		0.061
		(0.052)		(0.062)		(0.066)
Family Size		0.013		0.020		−0.051
		(0.050)		(0.059)		(0.064)
Lives with Partner (Base = Yes)		−0.019		0.058		−0.076
		(0.059)		(0.075)		(0.078)
Marital Status (Base: Married)
Separated		0.134**		−0.184		−0.203
		(0.063)		(0.233)		(0.264)
Divorced		0.030		−0.083		−0.032
		(0.038)		(0.061)		(0.058)
Widow or widower		−0.049		−0.029		−0.015
		(0.057)		(0.071)		(0.068)
Never Married		0.018		−0.047		−0.083
		(0.032)		(0.054)		(0.054)
Education (Base: Junior College & College)
Primary School		0.009		−0.298*		−0.196
		(0.159)		(0.170)		(0.149)
Junior & Senior High School		0.107		−0.111		−0.166
		(0.145)		(0.150)		(0.134)
University		0.145		−0.068		−0.157
		(0.144)		(0.149)		(0.133)
Other		0.169		−0.089		−0.142
		(0.146)		(0.153)		(0.137)
Not started		0.248		0.181		0.079
		(0.154)		(0.168)		(0.136)
Housing Type (Base: Self-Owned)
Rental		0.019		0.012		−0.064**
		(0.023)		(0.033)		(0.032)
Cost Free		0.118***		0.329***		0.221***
		(0.045)		(0.060)		(0.073)
Unknown		0.202**		−0.702***
		(0.079)		(0.107)
Occupation (Base: Paid Work)
Unemployed		0.044		−0.040		0.005
		(0.060)		(0.102)		(0.104)
Student		−0.002		0.039		0.210
		(0.119)		(0.181)		(0.140)
Unpaid Work		−0.000		0.167***		0.167***
		(0.039)		(0.052)		(0.053)
Pensioner		0.007		0.069		0.116***
		(0.028)		(0.045)		(0.045)
Work Disability		−0.067		−0.027		0.072
		(0.055)		(0.069)		(0.066)
Something else		−0.250		0.015		−0.355
		(0.285)		(0.283)		(0.309)
Income in Euros (Base: 2001 - 2500 Euros)
No Income		0.010		−0.027		−0.032
		(0.062)		(0.091)		(0.080)
500 or less		−0.120		0.069		−0.061
		(0.160)		(0.164)		(0.172)
501 to 1000		0.051		0.042		−0.004
		(0.044)		(0.080)		(0.074)
1001 to 1500		0.016		−0.001		−0.001
		(0.049)		(0.079)		(0.077)
1501 to 2000		−0.034		−0.002		0.016
		(0.048)		(0.072)		(0.068)
2501 to 3000		−0.002		0.047		−0.032
		(0.048)		(0.073)		(0.078)
3001 to 3500		−0.020		0.028		−0.003
		(0.049)		(0.075)		(0.077)
3501 to 4000		0.047		−0.007		0.062
		(0.048)		(0.088)		(0.082)
4001 to 4500		0.055		0.032		0.291***
		(0.050)		(0.094)		(0.057)
4501 to 5000		−0.018		0.091		0.031
		(0.067)		(0.094)		(0.093)
5001 to 7500		0.045		−0.039		0.016
		(0.048)		(0.090)		(0.086)
More than 7500		0.003		0.095		0.095
		(0.079)		(0.128)		(0.118)
Don't Know		−0.009		0.015		0.013
		(0.035)		(0.057)		(0.055)
Constant	0.893***	0.651***	0.712***	0.661***	0.794***	1.074***
	(0.011)	(0.200)	(0.017)	(0.218)	(0.016)	(0.215)
Controls	No	Yes	No	Yes	No	Yes
Observations	1221	1221	1221	1221	1077	1077
R-squared	0.001	0.036	0.000	0.040	0.000	0.063

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 7. Trust in 2020 & Preventive Behavior.

	Stay home if any symptoms	Stay home if any symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]
Trust in 2020	0.011***	0.007**	0.013***	0.009*	0.005	0.000
	(0.003)	(0.003)	(0.004)	(0.005)	(0.004)	(0.005)
Male		−0.036**		−0.037*		−0.071***
		(0.015)		(0.021)		(0.021)
Age (Years)		0.002**		0.002*		0.001
		(0.001)		(0.001)		(0.001)
Children		−0.044**		−0.037		−0.019
		(0.020)		(0.030)		(0.027)
Family Size		0.022		0.015		0.008
		(0.018)		(0.028)		(0.025)
Lives with Partner (Base = Yes)		−0.014		0.015		−0.003
		(0.030)		(0.042)		(0.041)
Marital Status (Base: Married)
Separated		0.130***		−0.165		−0.255
		(0.038)		(0.192)		(0.199)
Divorced		0.033		−0.028		−0.047
		(0.026)		(0.042)		(0.042)
Widow or widower		−0.045		0.021		−0.024
		(0.038)		(0.047)		(0.044)
Never Married		0.021		−0.011		−0.071**
		(0.022)		(0.034)		(0.036)
Education (Base: Junior College & College)
Primary School		0.048		−0.137		−0.088
		(0.106)		(0.109)		(0.098)
Junior & Senior High School		0.151		−0.033		−0.092
		(0.092)		(0.091)		(0.084)
University		0.176*		−0.019		−0.093
		(0.092)		(0.090)		(0.083)
Other		0.215**		0.020		−0.063
		(0.092)		(0.093)		(0.085)
Not started		0.065		0.180		−0.000
		(0.176)		(0.153)		(0.166)
Housing Type (Base: Self-Owned)
Rental		0.013		0.002		−0.058**
		(0.016)		(0.024)		(0.024)
Cost Free		0.048		−0.013		−0.067
		(0.065)		(0.122)		(0.135)
Unknown		0.135***		−0.165		0.220***
		(0.033)		(0.262)		(0.047)
Occupation (Base: Paid Work)
Unemployed		0.007		−0.054		−0.050
		(0.043)		(0.068)		(0.067)
Student		0.017		0.008		0.093
		(0.046)		(0.060)		(0.060)
Unpaid Work		0.001		0.058		0.062
		(0.030)		(0.041)		(0.039)
Pensioner		−0.009		0.035		0.044
		(0.022)		(0.034)		(0.033)
Work Disability		−0.009		−0.002		0.036
		(0.038)		(0.053)		(0.049)
Something else		−0.211		−0.111		−0.150
		(0.152)		(0.164)		(0.179)
Income in Euros (Base: 2001 - 2500 Euros)
No Income		−0.036		−0.058		−0.087
		(0.046)		(0.061)		(0.059)
500 or less		−0.005		−0.002		0.043
		(0.071)		(0.103)		(0.087)
501 to 1000		0.021		0.035		−0.004
		(0.039)		(0.061)		(0.057)
1001 to 1500		0.007		−0.046		0.008
		(0.036)		(0.058)		(0.054)
1501 to 2000		−0.013		−0.011		−0.024
		(0.035)		(0.053)		(0.052)
2501 to 3000		0.012		−0.015		−0.054
		(0.034)		(0.054)		(0.056)
3001 to 3500		−0.008		−0.028		−0.025
		(0.035)		(0.054)		(0.056)
3501 to 4000		0.056*		0.006		0.003
		(0.032)		(0.060)		(0.059)
4001 to 4500		0.000		0.020		0.141***
		(0.044)		(0.067)		(0.054)
4501 to 5000		0.008		−0.002		0.010
		(0.046)		(0.072)		(0.070)
5001 to 7500		0.014		−0.066		−0.011
		(0.037)		(0.063)		(0.060)
More than 7500		−0.009		−0.050		−0.008
		(0.065)		(0.106)		(0.100)
Don't Know		−0.023		−0.033		−0.022
		(0.027)		(0.041)		(0.040)
Constant	0.823***	0.568***	0.623***	0.578***	0.765***	0.897***
	(0.021)	(0.116)	(0.029)	(0.132)	(0.028)	(0.127)
Controls	No	Yes	No	Yes	No	Yes
Observations	2240	2240	2239	2239	1936	1936
R-squared	0.007	0.042	0.004	0.030	0.001	0.040

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 8. Trust in 2019 & Preventive Behavior.

	Stay home if any symptoms	Stay home if any symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]
Trust in 2019	0.011***	0.007**	0.014***	0.009*	0.007*	0.004
	(0.003)	(0.003)	(0.004)	(0.005)	(0.004)	(0.004)
Male		−0.042***		−0.043**		−0.070***
		(0.015)		(0.021)		(0.021)
Age (Years)		0.002**		0.002*		0.001
		(0.001)		(0.001)		(0.001)
Children		−0.045**		−0.034		−0.018
		(0.020)		(0.030)		(0.027)
Family Size		0.023		0.012		0.006
		(0.018)		(0.028)		(0.025)
Lives with Partner (Base = Yes)		−0.008		0.019		0.002
		(0.030)		(0.042)		(0.041)
Marital Status (Base: Married)
Separated		0.127***		−0.169		−0.247
		(0.038)		(0.192)		(0.197)
Divorced		0.036		−0.028		−0.053
		(0.027)		(0.042)		(0.042)
Widow or widower		−0.043		0.026		−0.028
		(0.038)		(0.047)		(0.044)
Never Married		0.022		−0.004		−0.071*
		(0.023)		(0.035)		(0.036)
Education (Base: Junior College & College)
Primary School		0.004		−0.177*		−0.132
		(0.104)		(0.106)		(0.095)
Junior & Senior High School		0.131		−0.048		−0.101
		(0.090)		(0.087)		(0.080)
University		0.156*		−0.029		−0.100
		(0.089)		(0.087)		(0.079)
Other		0.196**		0.008		−0.075
		(0.090)		(0.090)		(0.082)
Not started		0.054		0.171		−0.008
		(0.174)		(0.150)		(0.167)
Housing Type (Base: Self-Owned)
Rental		−0.001		−0.007		−0.061***
		(0.017)		(0.024)		(0.024)
Cost Free		0.046		−0.002		−0.051
		(0.061)		(0.116)		(0.124)
Unknown		0.132***		−0.161		0.218***
		(0.033)		(0.267)		(0.047)
Occupation (Base: Paid Work)
Unemployed		0.006		−0.030		−0.062
		(0.045)		(0.068)		(0.068)
Student		0.014		−0.005		0.091
		(0.047)		(0.061)		(0.060)
Unpaid Work		0.004		0.064		0.073*
		(0.030)		(0.040)		(0.038)
Pensioner		−0.004		0.039		0.050
		(0.022)		(0.033)		(0.033)
Work Disability		−0.015		0.013		0.044
		(0.040)		(0.053)		(0.048)
Something else		−0.193		−0.202		−0.141
		(0.158)		(0.159)		(0.182)
Income in Euros (Base: 2001 - 2500 Euros)
No Income		−0.037		−0.063		−0.097*
		(0.046)		(0.061)		(0.058)
500 or less		0.007		0.014		0.037
		(0.069)		(0.099)		(0.084)
501 to 1000		0.015		0.030		−0.018
		(0.040)		(0.062)		(0.057)
1001 to 1500		0.009		−0.050		−0.005
		(0.036)		(0.058)		(0.054)
1501 to 2000		−0.022		−0.027		−0.035
		(0.036)		(0.053)		(0.051)
2501 to 3000		0.007		−0.023		−0.063
		(0.035)		(0.055)		(0.056)
3001 to 3500		−0.007		−0.026		−0.064
		(0.035)		(0.055)		(0.056)
3501 to 4000		0.042		−0.003		−0.018
		(0.034)		(0.060)		(0.059)
4001 to 4500		−0.000		0.011		0.122**
		(0.044)		(0.067)		(0.054)
4501 to 5000		0.012		−0.001		−0.002
		(0.046)		(0.071)		(0.069)
5001 to 7500		0.018		−0.070		−0.028
		(0.037)		(0.063)		(0.060)
More than 7500		−0.002		−0.047		−0.022
		(0.066)		(0.105)		(0.100)
Don't Know		−0.024		−0.037		−0.038
		(0.027)		(0.041)		(0.039)
Constant	0.820***	0.589***	0.617***	0.583***	0.748***	0.902***
	(0.021)	(0.114)	(0.028)	(0.129)	(0.028)	(0.124)
Controls	No	Yes	No	Yes	No	Yes
Observations	2244	2244	2243	2243	1942	1942
R-squared	0.006	0.043	0.005	0.034	0.002	0.043

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 9. Trust in 2015 & Preventive Behavior.

	Stay home if any symptoms	Stay home if any symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]
Trust in 2015	0.012***	0.008**	0.014***	0.012**	0.002	0.001
	(0.004)	(0.004)	(0.005)	(0.005)	(0.005)	(0.005)
Male		−0.031*		−0.030		−0.062**
		(0.016)		(0.024)		(0.024)
Age (Years)		0.001*		0.001		−0.000
		(0.001)		(0.001)		(0.001)
Children		−0.043		−0.000		−0.014
		(0.035)		(0.050)		(0.045)
Family Size		0.028		−0.017		0.005
		(0.033)		(0.049)		(0.043)
Lives with Partner (Base = Yes)		0.017		0.031		0.029
		(0.043)		(0.061)		(0.057)
Marital Status (Base: Married)
Separated		0.089**		−0.113		−0.204
		(0.040)		(0.191)		(0.213)
Divorced		0.013		−0.081		−0.056
		(0.031)		(0.049)		(0.048)
Widow or widower		−0.048		0.003		−0.069
		(0.045)		(0.057)		(0.054)
Never Married		0.014		−0.022		−0.108**
		(0.025)		(0.042)		(0.044)
Education (Base: Junior College & College)
Primary School		−0.137		−0.354***		−0.166
		(0.114)		(0.119)		(0.104)
Junior & Senior High School		0.019		−0.177*		−0.148*
		(0.095)		(0.097)		(0.088)
University		0.042		−0.158*		−0.165*
		(0.094)		(0.096)		(0.088)
Other		0.074		−0.133		−0.155*
		(0.095)		(0.099)		(0.091)
Not started		−0.029		−0.055		0.088
		(0.205)		(0.208)		(0.096)
Housing Type (Base: Self-Owned)
Rental		0.018		0.008		−0.053**
		(0.018)		(0.027)		(0.027)
Cost Free		0.003		0.027		−0.064
		(0.083)		(0.138)		(0.160)
Unknown		0.111***		−0.361		0.202***
		(0.042)		(0.277)		(0.073)
Occupation (Base: Paid Work)
Unemployed		0.021		−0.078		−0.071
		(0.049)		(0.081)		(0.081)
Student		0.029		−0.028		0.143
		(0.061)		(0.098)		(0.091)
Unpaid Work		0.013		0.108**		0.124***
		(0.032)		(0.045)		(0.043)
Pensioner		−0.005		0.058		0.080**
		(0.024)		(0.038)		(0.038)
Work Disability		−0.031		−0.019		0.064
		(0.044)		(0.061)		(0.054)
Something else		−0.107		−0.079		−0.296
		(0.185)		(0.228)		(0.252)
Income in Euros (Base: 2001 - 2500 Euros)
No Income		−0.019		−0.005		−0.079
		(0.049)		(0.071)		(0.066)
500 or less		−0.013		0.060		0.019
		(0.094)		(0.121)		(0.106)
501 to 1000		0.016		0.024		−0.031
		(0.044)		(0.071)		(0.064)
1001 to 1500		−0.002		−0.010		−0.033
		(0.040)		(0.064)		(0.062)
1501 to 2000		−0.023		−0.022		−0.039
		(0.040)		(0.061)		(0.058)
2501 to 3000		0.009		0.011		−0.077
		(0.038)		(0.061)		(0.064)
3001 to 3500		0.008		0.025		−0.036
		(0.037)		(0.062)		(0.063)
3501 to 4000		0.056		0.007		−0.001
		(0.035)		(0.068)		(0.065)
4001 to 4500		0.033		0.037		0.149***
		(0.042)		(0.075)		(0.057)
4501 to 5000		−0.002		0.030		−0.017
		(0.052)		(0.078)		(0.079)
5001 to 7500		0.019		−0.081		−0.038
		(0.040)		(0.074)		(0.069)
More than 7500		−0.036		0.005		0.083
		(0.082)		(0.119)		(0.094)
Don't Know		−0.024		−0.014		−0.044
		(0.029)		(0.046)		(0.044)
Constant	0.828***	0.696***	0.624***	0.760***	0.786***	1.025***
	(0.024)	(0.135)	(0.033)	(0.161)	(0.031)	(0.152)
Controls	No	Yes	No	Yes	No	Yes
Observations	1731	1731	1731	1731	1504	1504
R-squared	0.008	0.034	0.005	0.034	0.000	0.049

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 10. Trust in 2010 & Preventive Behavior.

	Stay home if any symptoms	Stay home if any symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]
Trust in 2010	0.013***	0.010**	0.018***	0.016***	0.006	0.002
	(0.004)	(0.004)	(0.006)	(0.006)	(0.006)	(0.006)
Male		−0.026		−0.027		−0.081***
		(0.019)		(0.030)		(0.029)
Age (Years)		0.002		0.001		−0.001
		(0.001)		(0.002)		(0.002)
Children		−0.023		−0.029		0.059
		(0.052)		(0.061)		(0.066)
Family Size		0.010		0.016		−0.050
		(0.050)		(0.058)		(0.064)
Lives with Partner (Base = Yes)		−0.027		0.048		−0.069
		(0.059)		(0.075)		(0.078)
Marital Status (Base: Married)
Separated		0.164***		−0.136		−0.206
		(0.061)		(0.221)		(0.261)
Divorced		0.030		−0.077		−0.031
		(0.037)		(0.061)		(0.058)
Widow or widower		−0.047		−0.032		−0.030
		(0.057)		(0.072)		(0.068)
Never Married		0.026		−0.031		−0.079
		(0.031)		(0.054)		(0.055)
Education (Base: Junior College & College)
Primary School		0.000		−0.306*		−0.200
		(0.155)		(0.166)		(0.147)
Junior & Senior High School		0.101		−0.114		−0.161
		(0.141)		(0.146)		(0.132)
University		0.143		−0.070		−0.152
		(0.140)		(0.145)		(0.131)
Other		0.158		−0.095		−0.130
		(0.141)		(0.150)		(0.135)
Not started		0.210		0.138		0.092
		(0.152)		(0.174)		(0.134)
Housing Type (Base: Self-Owned)
Rental		0.026		0.027		−0.055*
		(0.023)		(0.032)		(0.033)
Cost Free		0.104***		0.303***		0.211***
		(0.040)		(0.058)		(0.075)
Unknown		0.179**		−0.749***
		(0.078)		(0.106)
Occupation (Base: Paid Work)
Unemployed		0.041		−0.042		0.003
		(0.059)		(0.101)		(0.105)
Student		−0.005		0.032		0.206
		(0.122)		(0.187)		(0.139)
Unpaid Work		0.004		0.171***		0.168***
		(0.037)		(0.052)		(0.054)
Pensioner		0.002		0.072		0.120***
		(0.028)		(0.045)		(0.045)
Work Disability		−0.069		−0.010		0.085
		(0.057)		(0.072)		(0.066)
Something else		−0.246		0.024		−0.355
		(0.279)		(0.277)		(0.298)
Income in Euros (Base: 2001 - 2500 Euros)
No Income		0.010		−0.028		−0.032
		(0.062)		(0.091)		(0.081)
500 or less		−0.102		0.098		−0.053
		(0.159)		(0.173)		(0.173)
501 to 1000		0.055		0.044		−0.008
		(0.045)		(0.081)		(0.074)
1001 to 1500		0.020		0.000		−0.003
		(0.049)		(0.080)		(0.077)
1501 to 2000		−0.011		0.017		0.015
		(0.046)		(0.071)		(0.069)
2501 to 3000		−0.005		0.045		−0.036
		(0.048)		(0.073)		(0.078)
3001 to 3500		−0.027		0.029		0.009
		(0.049)		(0.075)		(0.077)
3501 to 4000		0.044		−0.010		0.064
		(0.048)		(0.087)		(0.082)
4001 to 4500		0.046		0.021		0.291***
		(0.051)		(0.094)		(0.057)
4501 to 5000		−0.020		0.089		0.033
		(0.066)		(0.093)		(0.093)
5001 to 7500		0.036		−0.056		0.009
		(0.048)		(0.091)		(0.087)
More than 7500		−0.006		0.083		0.094
		(0.079)		(0.128)		(0.119)
Don't Know		−0.004		0.023		0.017
		(0.035)		(0.057)		(0.055)
Constant	0.823***	0.597***	0.610***	0.571***	0.760***	1.053***
	(0.028)	(0.199)	(0.040)	(0.215)	(0.039)	(0.217)
Controls	No	Yes	No	Yes	No	Yes
Observations	1206	1206	1206	1206	1064	1064
R-squared	0.009	0.041	0.007	0.046	0.001	0.062

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 11. Pro-sociality & Covid-19 Preventive Measures

(Using different controls to check the robustness of estimates).

	Stay home if any symptoms	Stay home if no symptoms	Stay home if govt. mandates
Regression#	[1]	[2]	[3]
Panel A: Prosocial Behavior in 2020 & Preventive Behavior
Donated in 2020	0.033**	0.036*	0.006
	(0.013)	(0.020)	(0.019)
Controls	Yes	Yes	Yes
Observations	2287	2286	1980
R-squared	0.021	0.023	0.024
Panel B: Prosocial Behavior in 2019 & Preventive Behavior
Donated in 2019	0.038***	−0.010	0.021
	(0.014)	(0.021)	(0.020)
Controls	Yes	Yes	Yes
Observations	2244	2243	1942
R-squared	0.020	0.022	0.024
Panel C: Prosocial Behavior in 2015 & Preventive Behavior
Donated in 2015	0.037**	0.021	0.008
	(0.015)	(0.024)	(0.022)
Controls	Yes	Yes	Yes
Observations	1769	1769	1533
R-squared	0.010	0.016	0.024
Panel D: Prosocial Behavior in 2010 & Preventive Behavior
Donated in 2010	0.016	0.002	−0.003
	(0.017)	(0.027)	(0.025)
Controls	Yes	Yes	Yes
Observations	1221	1221	1077
R-squared	0.011	0.016	0.026

Controls include gender, age, children, family size and marital status.

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 12. Trust & Covid-19 Preventive Measures

(Using different controls to check the robustness of estimates)

	Stay home if any symptoms	Stay home if no symptoms	Stay home if govt. mandates
Regression#	[1]	[2]	[3]
Panel A: Effect of Trust in 2020 on Preventive Behavior
Trust in 2020	0.010***	0.010**	0.004
	(0.003)	(0.004)	(0.004)
Controls	Yes	Yes	Yes
Observations	2240	2239	1936
R-squared	0.022	0.022	0.024
Panel B: Effect of Trust in 2019 on Preventive Behavior
Trust in 2019	0.010***	0.011**	0.006
	(0.003)	(0.004)	(0.004)
Controls	Yes	Yes	Yes
Observations	2244	2243	1942
R-squared	0.022	0.025	0.025
Panel C: Effect of Trust in 2015 on Preventive Behavior
Trust in 2015	0.012***	0.013***	0.002
	(0.004)	(0.005)	(0.005)
Controls	Yes	Yes	Yes
Observations	1731	1731	1504
R-squared	0.014	0.019	0.023
Panel D: Effect of Trust in 2010 on Preventive Behavior
Trust in 2010	0.013***	0.017***	0.006
	(0.004)	(0.006)	(0.006)
Controls	Yes	Yes	Yes
Observations	1206	1206	1064
R-squared	0.018	0.021	0.025

Controls include gender, age, children, family size and marital status.

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 13. Pro-sociality & Stay Home in Case of Symptoms (Identical Estimates).

	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms
Regression#	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]
Donated in 2010	0.020				0.005
	(0.017)				(0.019)
Donated in 2015		0.040**				0.031*
		(0.017)				(0.018)
Donated in 2019			0.045***				0.036**
			(0.017)				(0.018)
Donated in 2020				0.038**				0.019
				(0.017)				(0.018)
Male					−0.032	−0.030	−0.030	−0.030
					(0.020)	(0.020)	(0.020)	(0.020)
Age (Years)					0.002	0.002	0.002	0.002
					(0.001)	(0.001)	(0.001)	(0.001)
Children					−0.029	−0.032	−0.031	−0.030
					(0.052)	(0.052)	(0.052)	(0.053)
Family Size					0.013	0.016	0.014	0.014
					(0.050)	(0.050)	(0.050)	(0.051)
Lives with Partner (Base = Yes)					−0.019	−0.019	−0.017	−0.019
					(0.059)	(0.059)	(0.059)	(0.060)
Marital Status (Base: Married)
Separated					0.134**	0.132**	0.137**	0.134**
					(0.063)	(0.063)	(0.065)	(0.063)
Divorced					0.030	0.035	0.030	0.030
					(0.038)	(0.038)	(0.038)	(0.038)
Widow or widower					−0.049	−0.045	−0.050	−0.048
					(0.057)	(0.057)	(0.057)	(0.057)
Never Married					0.018	0.018	0.015	0.017
					(0.032)	(0.032)	(0.032)	(0.032)
Education (Base: Junior College & College)
Primary School					0.009	0.010	0.022	0.013
					(0.159)	(0.158)	(0.159)	(0.158)
Junior & Senior High School					0.107	0.108	0.117	0.109
					(0.145)	(0.143)	(0.145)	(0.143)
University					0.145	0.145	0.153	0.145
					(0.144)	(0.143)	(0.144)	(0.143)
Other					0.169	0.166	0.170	0.166
					(0.146)	(0.144)	(0.146)	(0.144)
Not started					0.248	0.249*	0.260*	0.246
					(0.154)	(0.150)	(0.152)	(0.151)
Housing Type (Base: Self-Owned)
Rental					0.019	0.022	0.021	0.020
					(0.023)	(0.023)	(0.023)	(0.023)
Cost Free					0.118***	0.128***	0.118**	0.124***
					(0.045)	(0.044)	(0.049)	(0.044)
Unknown					0.202**	0.210***	0.205**	0.205**
					(0.079)	(0.080)	(0.080)	(0.080)
Occupation (Base: Paid Work)
Unemployed					0.044	0.042	0.043	0.045
					(0.060)	(0.060)	(0.061)	(0.060)
Student					−0.002	0.001	0.002	−0.002
					(0.119)	(0.119)	(0.117)	(0.118)
Unpaid Work					−0.000	0.000	−0.001	−0.002
					(0.039)	(0.039)	(0.039)	(0.039)
Pensioner					0.007	0.008	0.008	0.007
					(0.028)	(0.028)	(0.028)	(0.028)
Work Disability					−0.067	−0.068	−0.068	−0.065
					(0.055)	(0.055)	(0.055)	(0.055)
Something else					−0.250	−0.242	−0.241	−0.244
					(0.285)	(0.286)	(0.284)	(0.285)
Income in Euros (Base: 2001 - 2500 Euros)
No Income					0.010	0.013	0.015	0.010
					(0.062)	(0.062)	(0.062)	(0.062)
500 or less					−0.120	−0.117	−0.117	−0.119
					(0.160)	(0.161)	(0.159)	(0.161)
501 to 1000					0.051	0.051	0.051	0.050
					(0.044)	(0.044)	(0.044)	(0.044)
1001 to 1500					0.016	0.019	0.020	0.017
					(0.049)	(0.048)	(0.049)	(0.048)
1501 to 2000					−0.034	−0.033	−0.033	−0.034
					(0.048)	(0.048)	(0.048)	(0.048)
2501 to 3000					−0.002	0.001	0.002	−0.002
					(0.048)	(0.048)	(0.048)	(0.048)
3001 to 3500					−0.020	−0.021	−0.021	−0.023
					(0.049)	(0.048)	(0.048)	(0.049)
3501 to 4000					0.047	0.046	0.048	0.047
					(0.048)	(0.048)	(0.048)	(0.048)
4001 to 4500					0.055	0.059	0.057	0.056
					(0.050)	(0.050)	(0.050)	(0.050)
4501 to 5000					−0.018	−0.017	−0.015	−0.018
					(0.067)	(0.067)	(0.067)	(0.067)
5001 to 7500					0.045	0.044	0.045	0.044
					(0.048)	(0.048)	(0.048)	(0.048)
More than 7500					0.003	−0.002	−0.004	−0.001
					(0.079)	(0.078)	(0.078)	(0.079)
Don't Know					−0.009	−0.007	−0.006	−0.008
					(0.035)	(0.035)	(0.035)	(0.035)
Constant	0.893***	0.889***	0.887***	0.887***	0.651***	0.640***	0.637***	0.649***
	(0.011)	(0.011)	(0.011)	(0.011)	(0.200)	(0.199)	(0.201)	(0.199)
Controls	No	No	No	No	Yes	Yes	Yes	Yes
Observations	1221	1221	1221	1221	1221	1221	1221	1221
R-squared	0.001	0.004	0.005	0.004	0.036	0.038	0.038	0.036

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 14. Pro-sociality & Stay Home in Case of no Symptoms (Identical Estimates).

	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms
Regression#	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]
Donated in 2010	0.014				−0.006
	(0.026)				(0.027)
Donated in 2015		0.023				0.018
		(0.028)				(0.029)
Donated in 2019			0.015				0.011
			(0.028)				(0.029)
Donated in 2020				0.046*				0.029
				(0.027)				(0.028)
Male					−0.026	−0.024	−0.025	−0.023
					(0.030)	(0.030)	(0.030)	(0.030)
Age (Years)					0.001	0.001	0.001	0.001
					(0.002)	(0.002)	(0.002)	(0.002)
Children					−0.037	−0.040	−0.039	−0.041
					(0.062)	(0.062)	(0.062)	(0.062)
Family Size					0.020	0.022	0.021	0.023
					(0.059)	(0.059)	(0.059)	(0.059)
Lives with Partner (Base = Yes)					0.058	0.059	0.059	0.059
					(0.075)	(0.075)	(0.075)	(0.075)
Marital Status (Base: Married)
Separated					−0.184	−0.184	−0.182	−0.182
					(0.233)	(0.232)	(0.232)	(0.232)
Divorced					−0.083	−0.080	−0.083	−0.083
					(0.061)	(0.061)	(0.061)	(0.061)
Widow or widower					−0.029	−0.027	−0.030	−0.028
					(0.071)	(0.071)	(0.071)	(0.071)
Never Married					−0.047	−0.048	−0.049	−0.050
					(0.054)	(0.054)	(0.054)	(0.054)
Education (Base: Junior College & College)
Primary School					−0.298*	−0.295*	−0.293*	−0.289*
					(0.170)	(0.168)	(0.169)	(0.167)
Junior & Senior High School					−0.111	−0.110	−0.108	−0.108
					(0.150)	(0.148)	(0.150)	(0.147)
University					−0.068	−0.068	−0.066	−0.068
					(0.149)	(0.148)	(0.149)	(0.147)
Other					−0.089	−0.093	−0.090	−0.096
					(0.153)	(0.152)	(0.153)	(0.151)
Not started					0.181	0.181	0.184	0.177
					(0.168)	(0.166)	(0.168)	(0.164)
Housing Type (Base: Self-Owned)
Rental					0.012	0.014	0.013	0.013
					(0.033)	(0.033)	(0.033)	(0.032)
Cost Free					0.329***	0.333***	0.327***	0.335***
					(0.060)	(0.062)	(0.062)	(0.061)
Unknown					−0.702***	−0.696***	−0.700***	−0.695***
					(0.107)	(0.107)	(0.107)	(0.106)
Occupation (Base: Paid Work)
Unemployed					−0.040	−0.041	−0.040	−0.039
					(0.102)	(0.102)	(0.102)	(0.102)
Student					0.039	0.041	0.040	0.040
					(0.181)	(0.181)	(0.182)	(0.184)
Unpaid Work					0.167***	0.167***	0.167***	0.165***
					(0.052)	(0.052)	(0.052)	(0.052)
Pensioner					0.069	0.071	0.070	0.071
					(0.045)	(0.045)	(0.045)	(0.045)
Work Disability					−0.027	−0.027	−0.027	−0.023
					(0.069)	(0.069)	(0.070)	(0.070)
Something else					0.015	0.020	0.018	0.024
					(0.283)	(0.282)	(0.282)	(0.282)
Income in Euros (Base: 2001 - 2500 Euros)
No Income					−0.027	−0.027	−0.027	−0.028
					(0.091)	(0.090)	(0.090)	(0.090)
500 or less					0.069	0.069	0.069	0.068
					(0.164)	(0.164)	(0.163)	(0.163)
501 to 1000					0.042	0.042	0.042	0.041
					(0.080)	(0.080)	(0.080)	(0.080)
1001 to 1500					−0.001	−0.001	−0.001	−0.002
					(0.079)	(0.079)	(0.079)	(0.079)
1501 to 2000					−0.002	−0.002	−0.003	−0.004
					(0.072)	(0.072)	(0.072)	(0.072)
2501 to 3000					0.047	0.048	0.047	0.046
					(0.073)	(0.073)	(0.073)	(0.073)
3001 to 3500					0.028	0.026	0.027	0.023
					(0.075)	(0.075)	(0.075)	(0.075)
3501 to 4000					−0.007	−0.009	−0.008	−0.009
					(0.088)	(0.088)	(0.087)	(0.087)
4001 to 4500					0.032	0.033	0.031	0.031
					(0.094)	(0.093)	(0.094)	(0.094)
4501 to 5000					0.091	0.090	0.091	0.088
					(0.094)	(0.094)	(0.094)	(0.093)
5001 to 7500					−0.039	−0.040	−0.040	−0.043
					(0.090)	(0.091)	(0.090)	(0.091)
More than 7500					0.095	0.091	0.092	0.087
					(0.128)	(0.128)	(0.128)	(0.129)
Don't Know					0.015	0.015	0.015	0.014
					(0.057)	(0.057)	(0.057)	(0.056)
Constant	0.712***	0.711***	0.713***	0.702***	0.661***	0.655***	0.657***	0.659***
	(0.017)	(0.015)	(0.015)	(0.016)	(0.218)	(0.218)	(0.218)	(0.217)
Controls	No	No	No	No	Yes	Yes	Yes	Yes
Observations	1221	1221	1221	1221	1221	1221	1221	1221
R-squared	0.000	0.001	0.000	0.002	0.040	0.040	0.040	0.040

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 15. Pro-sociality & Stay Home if government mandates (Identical Estimates).

	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]
Donated in 2010	0.004				−0.013
	(0.025)				(0.026)
Donated in 2015		−0.003				−0.013
		(0.027)				(0.027)
Donated in 2019			0.024				0.018
			(0.027)				(0.027)
Donated in 2020				0.022				0.013
				(0.025)				(0.026)
Male					−0.077***	−0.077***	−0.075**	−0.075**
					(0.029)	(0.029)	(0.029)	(0.030)
Age (Years)					−0.000	−0.000	−0.001	−0.001
					(0.002)	(0.002)	(0.002)	(0.002)
Children					0.061	0.061	0.059	0.059
					(0.066)	(0.067)	(0.066)	(0.066)
Family Size					−0.051	−0.052	−0.050	−0.049
					(0.064)	(0.065)	(0.064)	(0.064)
Lives with Partner (Base = Yes)					−0.076	−0.075	−0.073	−0.074
					(0.078)	(0.078)	(0.078)	(0.078)
Marital Status (Base: Married)
Separated					−0.203	−0.201	−0.198	−0.200
					(0.264)	(0.263)	(0.262)	(0.263)
Divorced					−0.032	−0.034	−0.032	−0.033
					(0.058)	(0.058)	(0.058)	(0.058)
Widow or widower					−0.015	−0.018	−0.017	−0.017
					(0.068)	(0.068)	(0.067)	(0.068)
Never Married					−0.083	−0.084	−0.086	−0.085
					(0.054)	(0.054)	(0.054)	(0.054)
Education (Base: Junior College & College)
Primary School					−0.196	−0.194	−0.184	−0.189
					(0.149)	(0.148)	(0.148)	(0.148)
Junior & Senior High School					−0.166	−0.166	−0.159	−0.163
					(0.134)	(0.133)	(0.133)	(0.133)
University					−0.157	−0.157	−0.152	−0.156
					(0.133)	(0.133)	(0.132)	(0.132)
Other					−0.142	−0.143	−0.142	−0.145
					(0.137)	(0.137)	(0.136)	(0.136)
Not started					0.079	0.075	0.089	0.071
					(0.136)	(0.136)	(0.136)	(0.135)
Housing Type (Base: Self-Owned)
Rental					−0.064**	−0.065**	−0.062*	−0.063*
					(0.032)	(0.033)	(0.032)	(0.032)
Cost Free					0.221***	0.212***	0.214***	0.220***
					(0.073)	(0.076)	(0.078)	(0.076)
Unknown					–	–	–	–
Occupation (Base: Paid Work)
Unemployed					0.005	0.004	0.004	0.005
					(0.104)	(0.105)	(0.105)	(0.105)
Student					0.210	0.207	0.213	0.211
					(0.140)	(0.138)	(0.138)	(0.140)
Unpaid Work					0.167***	0.167***	0.168***	0.167***
					(0.053)	(0.053)	(0.053)	(0.053)
Pensioner					0.116***	0.115***	0.118***	0.117***
					(0.045)	(0.045)	(0.045)	(0.045)
Work Disability					0.072	0.072	0.072	0.074
					(0.066)	(0.066)	(0.066)	(0.066)
Something else					−0.355	−0.358	−0.350	−0.351
					(0.309)	(0.305)	(0.306)	(0.306)
Income in Euros (Base: 2001 - 2500 Euros)
No Income					−0.032	−0.034	−0.032	−0.034
					(0.080)	(0.080)	(0.080)	(0.080)
500 or less					−0.061	−0.063	−0.062	−0.063
					(0.172)	(0.172)	(0.171)	(0.172)
501 to 1000					−0.004	−0.004	−0.004	−0.004
					(0.074)	(0.074)	(0.074)	(0.074)
1001 to 1500					−0.001	−0.003	−0.001	−0.003
					(0.077)	(0.077)	(0.078)	(0.077)
1501 to 2000					0.016	0.015	0.015	0.015
					(0.068)	(0.068)	(0.068)	(0.068)
2501 to 3000					−0.032	−0.035	−0.033	−0.034
					(0.078)	(0.078)	(0.078)	(0.078)
3001 to 3500					−0.003	−0.004	−0.005	−0.007
					(0.077)	(0.076)	(0.077)	(0.077)
3501 to 4000					0.062	0.061	0.060	0.060
					(0.082)	(0.082)	(0.082)	(0.082)
4001 to 4500					0.291***	0.287***	0.289***	0.288***
					(0.057)	(0.057)	(0.056)	(0.056)
4501 to 5000					0.031	0.028	0.029	0.027
					(0.093)	(0.093)	(0.093)	(0.093)
5001 to 7500					0.016	0.016	0.015	0.014
					(0.086)	(0.086)	(0.086)	(0.086)
More than 7500					0.095	0.095	0.087	0.089
					(0.118)	(0.118)	(0.120)	(0.119)
Don't Know					0.013	0.012	0.012	0.012
					(0.055)	(0.055)	(0.055)	(0.055)
Constant	0.794***	0.797***	0.789***	0.788***	1.074***	1.079***	1.066***	1.071***
	(0.016)	(0.015)	(0.015)	(0.015)	(0.215)	(0.215)	(0.214)	(0.214)
Controls	No	No	No	No	Yes	Yes	Yes	Yes
Observations	1077	1077	1077	1077	1077	1077	1077	1077
R-squared	0.000	0.000	0.001	0.001	0.063	0.063	0.063	0.063

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 16. Trust & Stay Home in Case of Symptoms (Identical Estimates).

	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms	Stay home if any symptoms
Regression#	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]
Trust in 2010	0.012***				0.009**
	(0.004)				(0.004)
Trust in 2015		0.013***				0.010**
		(0.004)				(0.004)
Trust in 2019			0.013***				0.010**
			(0.004)				(0.004)
Trust in 2020				0.012***				0.010**
				(0.004)				(0.004)
Male					−0.024	−0.024	−0.024	−0.023
					(0.019)	(0.019)	(0.019)	(0.019)
Age (Years)					0.001	0.001	0.001	0.001
					(0.001)	(0.001)	(0.001)	(0.001)
Children					−0.021	−0.019	−0.021	−0.023
					(0.052)	(0.052)	(0.052)	(0.052)
Family Size					0.006	0.004	0.006	0.007
					(0.050)	(0.050)	(0.050)	(0.050)
Lives with Partner (Base = Yes)			−0.033	−0.034	−0.034	−0.031
					(0.059)	(0.059)	(0.059)	(0.059)
Marital Status (Base: Married)
Separated					0.167**	0.166**	0.179***	0.178***
					(0.065)	(0.065)	(0.067)	(0.066)
Divorced					0.027	0.025	0.028	0.026
					(0.037)	(0.037)	(0.037)	(0.037)
Widow or widower					−0.052	−0.048	−0.050	−0.055
					(0.058)	(0.058)	(0.058)	(0.058)
Never Married					0.038	0.036	0.037	0.034
					(0.029)	(0.030)	(0.029)	(0.030)
Education (Base: Junior College & College)
Primary School					0.046	0.050	0.050	0.051
					(0.154)	(0.152)	(0.154)	(0.155)
Junior & Senior High School					0.108	0.111	0.111	0.113
					(0.141)	(0.139)	(0.141)	(0.141)
University					0.150	0.153	0.151	0.154
					(0.140)	(0.138)	(0.140)	(0.140)
Other					0.166	0.168	0.167	0.171
					(0.141)	(0.139)	(0.141)	(0.142)
Not started					0.218	0.249*	0.220	0.228
					(0.151)	(0.150)	(0.151)	(0.150)
Housing Type (Base: Self-Owned)
Rental					0.038*	0.038*	0.039*	0.040*
					(0.022)	(0.022)	(0.022)	(0.022)
Cost Free					0.095***	0.087**	0.084**	0.091**
					(0.036)	(0.038)	(0.036)	(0.036)
Unknown					0.155**	0.209***	0.166**	0.168**
					(0.076)	(0.079)	(0.076)	(0.077)
Occupation (Base: Paid Work)
Unemployed
			0.029	0.026	0.028	0.027
					(0.059)	(0.060)	(0.060)	(0.060)
Student					−0.056	−0.055	−0.058	−0.057
					(0.133)	(0.127)	(0.133)	(0.135)
Unpaid Work					−0.004	−0.003	−0.002	−0.001
					(0.039)	(0.039)	(0.039)	(0.039)
Pensioner					0.005	0.005	0.005	0.006
					(0.028)	(0.028)	(0.028)	(0.028)
Work Disability					−0.070	−0.063	−0.067	−0.065
					(0.056)	(0.056)	(0.056)	(0.056)
Something else					−0.259	−0.257	−0.257	−0.261
					(0.275)	(0.280)	(0.273)	(0.272)
Income in Euros (Base: 2001 - 2500 Euros)
No Income					−0.002	0.002	−0.004	−0.006
					(0.063)	(0.063)	(0.063)	(0.064)
500 or less					−0.116	−0.133	−0.118	−0.121
					(0.158)	(0.160)	(0.158)	(0.160)
501 to 1000					0.050	0.048	0.045	0.044
					(0.045)	(0.045)	(0.045)	(0.045)
1001 to 1500					0.011	0.013	0.012	0.011
					(0.048)	(0.048)	(0.048)	(0.049)
1501 to 2000					−0.010	−0.010	−0.009	−0.014
					(0.045)	(0.046)	(0.045)	(0.046)
2501 to 3000					0.006	0.003	0.006	−0.000
					(0.046)	(0.046)	(0.046)	(0.046)
3001 to 3500					−0.029	−0.027	−0.028	−0.031
					(0.049)	(0.048)	(0.048)	(0.049)
3501 to 4000					0.073*	0.072*	0.075*	0.069*
					(0.039)	(0.040)	(0.039)	(0.040)
4001 to 4500					0.035	0.034	0.035	0.032
					(0.050)	(0.050)	(0.049)	(0.050)
4501 to 5000					−0.031	−0.029	−0.029	−0.032
					(0.067)	(0.067)	(0.067)	(0.067)
5001 to 7500					0.027	0.024	0.028	0.024
					(0.048)	(0.048)	(0.048)	(0.048)
More than 7500					−0.021	−0.012	−0.021	−0.026
					(0.084)	(0.083)	(0.085)	(0.084)
Don't Know					−0.005	−0.006	−0.007	−0.009
					(0.034)	(0.034)	(0.034)	(0.034)
Constant	0.835***	0.829***	0.830***	0.834***	0.640***	0.642***	0.647***	0.646***
	(0.028)	(0.028)	(0.029)	(0.029)	(0.197)	(0.194)	(0.195)	(0.196)
Controls	No	No	No	No	Yes	Yes	Yes	Yes
Observations	1152	1152	1152	1152	1152	1152	1152	1152
R-squared	0.008	0.010	0.010	0.009	0.041	0.043	0.042	0.042

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 17. Trust & Stay Home in Case of no Symptoms (Identical Estimates).

	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms	Stay home if no symptoms
Regression#	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]
Trust in 2010	0.018***				0.017***
	(0.006)				(0.006)
Trust in 2015		0.014**				0.013**
		(0.006)				(0.006)
Trust in 2019			0.016***				0.013**
			(0.006)				(0.007)
Trust in 2020				0.009				0.006
				(0.006)				(0.006)
Male					−0.024	−0.024	−0.025	−0.025
					(0.030)	(0.030)	(0.030)	(0.030)
Age (Years)					0.000	0.000	0.000	0.000
					(0.002)	(0.002)	(0.002)	(0.002)
Children					−0.031	−0.028	−0.031	−0.033
					(0.061)	(0.062)	(0.062)	(0.062)
Family Size					0.015	0.013	0.015	0.017
					(0.058)	(0.059)	(0.059)	(0.059)
Lives with Partner (Base = Yes)			0.052	0.052	0.051	0.055
					(0.076)	(0.076)	(0.076)	(0.076)
Marital Status (Base: Married)
Separated					−0.127	−0.142	−0.123	−0.147
					(0.218)	(0.220)	(0.223)	(0.226)
Divorced					−0.090	−0.091	−0.088	−0.089
					(0.062)	(0.062)	(0.062)	(0.062)
Widow or widower					−0.040	−0.035	−0.038	−0.043
					(0.074)	(0.073)	(0.074)	(0.074)
Never Married					−0.023	−0.023	−0.022	−0.023
					(0.055)	(0.055)	(0.055)	(0.055)
Education (Base: Junior College & College)
Primary School					−0.257	−0.253	−0.253	−0.256
					(0.167)	(0.166)	(0.168)	(0.170)
Junior & Senior High School					−0.118	−0.113	−0.113	−0.112
					(0.146)	(0.145)	(0.146)	(0.148)
University					−0.071	−0.064	−0.067	−0.061
					(0.146)	(0.144)	(0.146)	(0.148)
Other					−0.089	−0.083	−0.085	−0.077
					(0.150)	(0.148)	(0.150)	(0.152)
Not started					0.146	0.184	0.148	0.151
					(0.173)	(0.170)	(0.172)	(0.170)
Housing Type (Base: Self-Owned)
Rental					0.036	0.033	0.035	0.032
					(0.033)	(0.033)	(0.033)	(0.033)
Cost Free					0.296***	0.293***	0.288***	0.308***
					(0.054)	(0.054)	(0.053)	(0.053)
Unknown					−0.771***	−0.688***	−0.744***	−0.731***
					(0.108)	(0.109)	(0.108)	(0.108)
Occupation (Base: Paid Work)
Unemployed				−0.048	−0.051	−0.050	−0.050
					(0.101)	(0.101)	(0.101)	(0.102)
Student					−0.060	−0.055	−0.060	−0.056
					(0.195)	(0.190)	(0.196)	(0.195)
Unpaid Work					0.168***	0.172***	0.173***	0.174***
					(0.054)	(0.054)	(0.054)	(0.054)
Pensioner					0.075	0.077*	0.076*	0.077*
					(0.046)	(0.046)	(0.046)	(0.046)
Work Disability					−0.007	−0.002	−0.007	−0.010
					(0.073)	(0.072)	(0.073)	(0.072)
Something else					0.023	0.023	0.024	0.017
					(0.273)	(0.266)	(0.275)	(0.275)
Income in Euros (Base: 2001 - 2500 Euros)
No Income					−0.048	−0.046	−0.053	−0.055
					(0.094)	(0.095)	(0.094)	(0.095)
500 or less					0.087	0.059	0.078	0.069
					(0.173)	(0.167)	(0.167)	(0.167)
501 to 1000					0.036	0.034	0.029	0.030
					(0.083)	(0.083)	(0.082)	(0.083)
1001 to 1500					−0.003	0.000	−0.002	−0.003
					(0.081)	(0.081)	(0.081)	(0.081)
1501 to 2000					0.032	0.032	0.032	0.029
					(0.072)	(0.072)	(0.072)	(0.073)
2501 to 3000					0.054	0.050	0.054	0.050
					(0.074)	(0.074)	(0.074)	(0.074)
3001 to 3500					0.032	0.037	0.035	0.037
					(0.076)	(0.076)	(0.076)	(0.076)
3501 to 4000					0.013	0.015	0.019	0.016
					(0.089)	(0.089)	(0.089)	(0.089)
4001 to 4500					0.013	0.017	0.017	0.021
					(0.094)	(0.094)	(0.094)	(0.094)
4501 to 5000					0.078	0.083	0.082	0.082
					(0.095)	(0.095)	(0.095)	(0.095)
5001 to 7500					−0.071	−0.071	−0.066	−0.064
					(0.092)	(0.092)	(0.092)	(0.092)
More than 7500					0.142	0.159	0.146	0.150
					(0.120)	(0.120)	(0.121)	(0.120)
Don't Know					0.020	0.019	0.018	0.017
					(0.057)	(0.057)	(0.057)	(0.057)
Constant	0.611***	0.635***	0.623***	0.664***	0.606***	0.629***	0.633***	0.653***
	(0.041)	(0.039)	(0.041)	(0.040)	(0.217)	(0.218)	(0.218)	(0.220)
Controls	No	No	No	No	Yes	Yes	Yes	Yes
Observations	1152	1152	1152	1152	1152	1152	1152	1152
R-squared	0.008	0.005	0.006	0.002	0.043	0.040	0.040	0.037

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.

Appendix 18. Trust & Stay Home if government mandates (Identical Estimates).

	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates	Stay home if govt. mandates
Regression#	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]
Trust in 2010	0.008				0.003
	(0.006)				(0.006)
Trust in 2015		0.007				0.005
		(0.006)				(0.006)
Trust n 2019			0.011*				0.007
			(0.006)				(0.006)
Trust in 2020				0.008				0.004
				(0.006)				(0.006)
Male					−0.069**	−0.069**	−0.069**	−0.069**
					(0.030)	(0.030)	(0.030)	(0.030)
Age (Years)					−0.001	−0.001	−0.001	−0.001
					(0.002)	(0.002)	(0.002)	(0.002)
Children					0.058	0.059	0.059	0.057
					(0.066)	(0.067)	(0.066)	(0.066)
Family Size					−0.053	−0.054	−0.054	−0.053
					(0.064)	(0.064)	(0.064)	(0.064)
Lives with Partner (Base = Yes)			−0.066	−0.067	−0.068	−0.066
					(0.079)	(0.079)	(0.079)	(0.079)
Marital Status (Base: Married)
Separated					−0.214	−0.210	−0.196	−0.208
					(0.264)	(0.262)	(0.262)	(0.264)
Divorced					−0.029	−0.030	−0.029	−0.029
					(0.059)	(0.059)	(0.058)	(0.059)
Widow or widower					−0.038	−0.035	−0.035	−0.038
					(0.071)	(0.071)	(0.070)	(0.071)
Never Married					−0.080	−0.082	−0.082	−0.082
					(0.056)	(0.056)	(0.056)	(0.056)
Education (Base: Junior College & College)
Primary School					−0.161	−0.159	−0.156	−0.158
					(0.146)	(0.145)	(0.144)	(0.147)
Junior & Senior High School					−0.161	−0.160	−0.159	−0.159
					(0.131)	(0.130)	(0.129)	(0.131)
University					−0.156	−0.156	−0.157	−0.155
					(0.130)	(0.129)	(0.128)	(0.130)
Other					−0.135	−0.135	−0.137	−0.133
					(0.135)	(0.133)	(0.132)	(0.135)
Not started					0.099	0.120	0.106	0.105
					(0.134)	(0.137)	(0.132)	(0.135)
Housing Type (Base: Self-Owned)
Rental					−0.055*	−0.054	−0.053	−0.054
					(0.034)	(0.033)	(0.034)	(0.034)
Cost Free					0.207***	0.203***	0.198***	0.206***
					(0.073)	(0.076)	(0.075)	(0.074)
Unknown
Occupation (Base: Paid Work)
Unemployed				0.002	0.001	0.002	0.002
					(0.104)	(0.104)	(0.105)	(0.104)
Student					0.189	0.191	0.189	0.189
					(0.160)	(0.160)	(0.161)	(0.161)
Unpaid Work					0.161***	0.161***	0.161***	0.162***
					(0.055)	(0.055)	(0.055)	(0.055)
Pensioner					0.114**	0.114**	0.114**	0.114**
					(0.046)	(0.046)	(0.045)	(0.046)
Work Disability					0.077	0.081	0.081	0.079
					(0.068)	(0.068)	(0.068)	(0.068)
Something else					−0.356	−0.354	−0.352	−0.356
					(0.299)	(0.297)	(0.295)	(0.299)
Income in Euros (Base: 2001 - 2500 Euros)
No Income					−0.025	−0.023	−0.026	−0.027
					(0.084)	(0.084)	(0.084)	(0.084)
500 or less					−0.052	−0.058	−0.049	−0.054
					(0.175)	(0.174)	(0.175)	(0.175)
501 to 1000					−0.002	−0.003	−0.005	−0.004
					(0.076)	(0.076)	(0.076)	(0.076)
1001 to 1500					−0.000	0.000	−0.001	−0.001
					(0.079)	(0.079)	(0.079)	(0.079)
1501 to 2000					0.020	0.019	0.020	0.018
					(0.070)	(0.070)	(0.070)	(0.070)
2501 to 3000					−0.050	−0.053	−0.051	−0.053
					(0.081)	(0.081)	(0.081)	(0.081)
3001 to 3500					0.008	0.007	0.005	0.006
					(0.080)	(0.079)	(0.080)	(0.079)
3501 to 4000					0.101	0.099	0.101	0.098
					(0.081)	(0.081)	(0.081)	(0.081)
4001 to 4500					0.289***	0.289***	0.287***	0.286***
					(0.058)	(0.057)	(0.057)	(0.058)
4501 to 5000					0.026	0.026	0.025	0.025
					(0.096)	(0.096)	(0.096)	(0.096)
5001 to 7500					0.004	0.001	0.002	0.002
					(0.088)	(0.088)	(0.088)	(0.088)
More than 7500					0.164	0.166	0.160	0.160
					(0.103)	(0.103)	(0.103)	(0.103)
Don't Know					0.019	0.019	0.018	0.017
					(0.056)	(0.056)	(0.056)	(0.056)
Constant	0.749***	0.755***	0.730***	0.749***	1.049***	1.041***	1.038***	1.049***
	(0.040)	(0.039)	(0.040)	(0.040)	(0.219)	(0.217)	(0.215)	(0.218)
Controls	No	No	No	No	Yes	Yes	Yes	Yes
Observations	1018	1018	1018	1018	1018	1018	1018	1018
R-squared	0.002	0.001	0.003	0.002	0.060	0.061	0.061	0.060

Robust standard errors clustered around individual IDs are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10.