2020 December 2020 December HOUSEHOLD SURVEY OVERVIEW 20-minute phone survey to capture impacts of COVID-19 on household well-being and policy uptake SAMPLING APPROACH, SAMPLE CAPTURED Sample size 1500 households Representativeness National representative sample of households by region and settlement size Method of interview Telephone interview (CATI) Fieldwork June 2nd to June 25th , 2020 Questionnaire design All indicators and questionnaire developed by the World Bank HOUSEHOLD SURVEY OVERVIEW Comparison of sample and population Age structure (using household roster) Population Survey Total Male Female Total Male Female 0–17 17.3 17.2 17.6 17.0 18–24 8.1 8.5 9.1 8.0 25–49 32.6 30.4 33.7 27.5 50–64 21.7 20.7 19.7 21.7 65+ 20.3 23.0 19.8 25.8 Education structure (using household roster) Population 18-64 Survey 18-64 Total Male Female Total Male Female Primary 15.0 11.6 9.8 13.3 Secondary, vocational 64.5 65.9 67.9 64.1 Tertiary 20.5 19.6 18.9 20.3 HOUSEHOLD SURVEY OVERVIEW Questionnaire Demographic Effects of COVID-19 and earthquake Employment and income Safety nets Knowledge about COVID-19 Change of behavior because of COVID-19 Mental health Children, school & health EDUCATION CITIZEN’S KNOWLEDGE AND BEHAVIOR TOWARD COVID PERCEPTIONS OF SUBJECTIVE GOVERNMENT WELLBEING RESPONSE EMPLOYMENT AND INCOME KEY MESSAGES Behavior and opinion Most Croatians (88%) know what to do to prevent the spread of COVID-19 • But women and richer households are more likely to act on it. A majority of people (77%) approved Government’s response to the crisis, with stronger support from women, older people, lower-income group, and urban residents • But men, young people, people in higher income bracket, or people with a job before the crisis hit are more likely to believe the government’s reaction to the current crisis is too extreme. KEY MESSAGES Employment and income COVID-19 has direct and major impacts to employment • One in four workers stopped working, particularly blue-collared workers. • 60 percent of work stoppage was due to COVID-19 business loss. • Even among those continued working, about 15 percent reduced their number of hours and experienced pay cut. A quarter of Croatian households had a reduction in income • Younger people and working people are more likely to see their income decline. • Richer households and households with working members experienced more % income drop. KEY MESSAGES Savings and subjective welfare A worrisome sign is the low rate of savings which make it more difficult for households to cushion the shock • 58% reported no savings. The situation is more difficult for poorer households, and households without any member working. • Even among household with savings, 76% of them would run out of savings in the next 6 months, particularly for poor and older people. Half of Croatian households experienced difficulties making enough money to pay for their current expenses • Poor households in rural areas, with older household head, or households without any members working suffered the most. KEY MESSAGES Access to healthcare While 90% households worrying about their health situation, access to health services is relatively high • On average, 40% households needed medical care during the pandemic. The rate is higher among women, lower income, and older people. • 85% households can get the medical care they needed. But the rate is significantly lower among young and richer people, probably because of the differences in types of treatment (urgency, etc.) KEY MESSAGES Access to education While most school offers some sorts of distant learning, access to online learning varies significantly across demographic groups • 88% students had access to online learning, the most advance and interactive distant learning. • The rate is significantly lower among girls, rural students, students from lower income households, and households with no working adults. In addition, distribution of distant learning equipment from school was uneven across groups. And the level is relatively low – only one in three students received a tablet from school • Interestingly, the poorest and the richest students had equal chance of receiving tablets from school, which may reflect the fact that richer schools were more equipped, and other schools targeted poor students well. KEY MESSAGES Access to education However, almost all students have access to tablets with internet (private) • The level of shared tablets within a households is higher among poor and rural households. Students required and received lots of help from another household members, often the non-working members • 84% Croatian students received support with schoolwork. • Younger and richer students are more likely to have help. KEY MESSAGES Access to education Despite all the support and equipment, distant learning took a toll on student’s mental health • One in four students were either unhappy or experienced problems with it. • Boys, younger students, students from households with no adults working, and especially students from higher income, were more likely to have trouble coping with distant learning. Parents were slightly worried about school reopening • The most concern is about the child getting infected. • 16% parents worried about school reopening. • Poor parents, and parents with high school students tend to worry more. CITIZEN’S KNOWLEDGE AND BEHAVIOR TOWARD COVID Knowledge about mitigation measures against the spread of COVID-19 is common, albeit at a lower rate among people in rural areas, and among those not working • On average, 88 % of households have received the information on how to reduce the risk. • The rate is higher in urban areas, and among working people. • Knowledge of COVID-19 does not vary by age or gender of respondents. Percentage of households receiving any information on how to reduce the risk of getting COVID-19 92 91 91 90 89 89 88 88 87 86 86 85 85 84 83 82 Working Not working Urban Rural All Pre-crisis employment status Location But changes in behavior vary significantly across population groups Women are significantly more likely to take action 100 87 89 90 85 76 79 76 80 71 70 67 60 50 40 30 20 10 0 Wearing mask More handwashing Social distancing Fewer/no visit families Male Female and friends Poorer households tend not to apply mitigation measures as much as better-off households 100 90 86 84 80 80 82 80 75 76 70 67 60 50 40 30 20 10 0 Wearing mask More handwashing Social distancing Fewer/no visit families and friends Bottom 40 income Top 20 income As of June 2020, nearly three quarters of Croatians had already practiced measures to reduce the spread of COVID-19 for more than a month Older people are more likely to apply these measures earlier than their younger cohorts • However, gender, location, or income level does not impact the time length practicing measures to reduce the risks of contracting COVID-19. Percentage of households practicing mitigation measures for more than one month 90 83 83 80 77 70 70 66 60 50 40 30 20 10 0 All Age 20-29 Age 30-39 Age 40-49 Age 50-64 PERCEPTIONS OF GOVERNMENT RESPONSE There is broad agreement of government policy responses to COVID-19, although nearly 15 percent of respondents considered the response too extreme Majority agreement with the response can be seen across all population groups, however, some differences by age/gender/income level. of respondents were supportive of government 77% response to COVID-19 in June 2020 consider insufficient 5% government response 3% had no reaction considered reaction too 15% extreme. Young, male, rich people tend to agree more. A majority of the population approves the Government’s crisis response policies, with stronger support from women, older people, lower- income group, and urban residents 84 82 82 81 80 80 79 79 78 76 76 75 74 74 74 72 72 72 70 68 66 Male Female Age 20- Age 30- Age 40- Age 50- Bottom Middle Top 20 Rural Urban 29 39 49 64 40 40 Gender Age Income level Location Meanwhile, men, young people, people in higher income bracket, or people with a job before the crisis hit are more likely to believe the government’s reaction to the current crisis is too extreme • However, it important to note that rural residents and people who did not work before the crisis tend to consider the government responses is insufficient although the number is relatively low. 30 25 25 21 20 20 19 18 16 15 13 14 12 11 10 10 5 0 Male Female Age 20- Age 30- Age 40- Age 50- Bottom Middle Top 20 Not Working 29 39 49 64 40 40 working Gender Age Income level Pre-crisis employment EMPLOYMENT AND INCOME Before the pandemic hit (Feb 2020), two third of household heads were employed Household heads who are male, live in urban, or in high-income bracket were significantly more likely to work. 100 93 90 81 80 80 78 72 70 65 65 60 53 50 40 30 20 10 0 Bottom 40 Middle 40 Top 20 Male Female Rural Urban All Income level Gender Location Pre-crisis job structure was very different across population groups Poor household heads were more likely to work in manufacture and construction sectors, while better-off people worked in public administration. The same pattern is observed among male and female household heads. 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Bottom 40 Middle 40 Top 20 Male Female All Income Gender Agriculture Manufacture, construction Transport, hotel, trade Finance, real estate Public administration Health, education, social work Pre-crisis occupation of household heads was split equally between white and blue collar, but female heads and richer household heads were much more likely to hold a white-collar job 100% 90% 80% 34 39 47 49 70% 57 65 60% 50% 40% 64 61 30% 50 52 20% 40 34 10% 0% Bottom 40 Middle 40 Top 20 Male Female All Income Gender White collar Blue collar Other Pre-crisis, nearly half of Croatian heads of households had a salaried job in private sector But lower-income and female heads were more likely to be salaried employees in public sector. 100% 5 8 12 14 13 13 90% 80% 70% 43 38 54 46 47 49 60% 50% 40% 30% 37 42 34 32 34 27 20% 10% 8 11 8 11 7 7 0% Bottom 40 Middle 40 Top 20 Male Female All Income Gender Self-employed Salaried (public sec) Salaried (private sec) Other As the pandemic spread, one in four workers* stopped working Low-income workers were hit the hardest. Similarly, blue-collar workers One in three stopped working by June suffered more 40% 40% 36% % working in Feb and stop working now 35% 34% 35% 32% 32% 31% 29% 30% 30% 24% 24% 25% 25% 21% 20% 20% 19% 22% 20% 16% 20% 18% 15% 15% 13% 14% 15% 10% 10% 5% 5% 0% 0% Feb mid-March mid-April early-Jun Feb mid-March mid-April early-Jun Bottom 40 Middle 40 Top 20 All White collar Blue collar 30% 26% 27% 25% However, while women were more likely 22% 22% 22% to stop working in March, work stoppage 20% was similar between men and women in 21% April. And in June, the pattern was 15% reversed 10% 5% 0% * household heads who were employed in February Feb mid-March mid-April early-Jun Male Female It is clear COVID-19 was the direct cause for two- third of work stoppage in March, and half of work stoppage in April March April Business went bankrupt or 64% 51% temporarily closed due to COVID-19 reasons Lost job due to less 6% 4% business Need to take care for kids 2% 4% at home from school 6% Ill or need to care for 5% ill relative 22% 36% Other In fact, 14 percent of firms have permanently closed, which contributes to work stoppage during the pandemic • The left side of the figure shows the share of firms that were confirmed to have closed since the pandemic was declared in March. • The right side uses a wider definition of closed firms: it includes the firms that closed since 2019 and also firms that could not be contacted during fieldwork and therefore are assumed to have closed. 16 14 12 10 8 6 4 2 0.1 0.3 0.0 0.0 0 % confirmed permanently closed since pademic % confirmed or ssumed permanently closed declared Croatia Small Medium Large For those workers stopped working in March and April, the impacts of COVID-19 were more severe for women and blue-collar workers 60 percent of women who stopped working in April reported their reason as business closure due to COVID-19, compared to 44 percent of men 80 68 % workers reported reason for not 70 60 60 60 50 44 working 40 30 20 8 10 3 0 0 0 Male Female Male Female Business went bankrupt or temporarily Need to take care for kids at home from closed due to COVID school March April Similarly, 57 percent of blue-collar workers reported COVID-reasons compared to 42 percent of white collars 80 70 70 59 57 60 50 42 40 30 20 10 4 4 3 0 0 White collar Blue collar White collar Blue collar Business went bankrupt or temporarily Need to take care for kids at home from closed due to COVID school March April Even among those continuing to work during the pandemic, the number of hours of work was cut significantly However, the trend appears to slowly revert to the pre-crisis level • 17% of those who continued to work in March reported a reduction in work hours. However, this number has gradually declined to 8 percent in June. • Female workers experienced a reduction in work hours more than men in the period right after COVID-19 was declared a pandemic (March and April). But the trend was reversed in June. • White-collar workers suffered the most in the immediate aftermath of COVID-19, but the reduction has been lessened over time. • Meanwhile, blue-collar workers have seen stable reduction of work hours over time. 25 20 20 19 17 15 15 14 13 11 11 11 11 10 10 8 8 7 5 5 0 Male Female White collar Blue collar All Gender Occupation March April June Compared to the same time last year, the number of people absent from work increased dramatically in March and April Number of people absent from work in Croatia in 2020, compared to 2019 600 500 400 Weeks 300 200 100 0 2019 2020 Share of employment (as % od Q4, 2019 employed) 35% 30% 25% 20% Weeks 15% 10% 5% 0% 2019 2020 Source: Eurostat 2020 The impacts of the pandemic also reflect at firm level A quarter of firms reported a decrease of weekly hours worked relative to before the outbreak 30 26 25 24 21 20 20 19 18 % of Firms 15 10 5 0 Croatia Small Medium Large Manuf. Services On average, it would take about 9 months before firms could return to their normal level of business. It would take even longer for large firms 12 11.6 10 9.3 9.6 9.1 9 8.7 8 Months 6 4 2 0 Croatia Small Medium Large Manuf. Services Source: Croatia Enterprise Survey 2020 Following a cut in work hours, workers experienced a reduction in salary Women tends to face a reduction in salary more than men, especially in April 16 15 % workers received partial or no 14 13 12 12 11 11 salary payments 10 8 8 6 4 2 0 March April All Male Female But interestingly, while blue-collar workers were more likely to stop working, those who continued to work did not appear to have more reduction in salaries than white-collar workers 14 12 12 12 11 11 10 10 9 8 6 4 2 0 March April All White collar Blue collar With higher work stoppage, reduction in work hours and salary, it is expected that household income would decline. In fact, one in four households reported so 24% of households report income drops. Households with younger household head tend to report a decline in income. Employment channel is clear for incomes***: • Households with employed members are most likely to report drops in income, 29% reported decreased income. • Among 34% of households reporting no working members living in the household, only 16% report deteriorating incomes. • However, gender of household head does not play a significant role in household income reduction. Percentage of households reporting declines in income since February 2020 7% 8% 11% 13% 6% 9% 14% 44% 58% 62% 69% 65% 75% 71% 39% 33% 29% 21% 24% 13% 16% <25 25-49 50-64 65+ Not working Working Age of household head Employment status in Feb All Reduced Same Increased Don't know Richer households, and households with at least one member working in February tend to have a larger decline in income All Employment status in Feb Working Not working Top 20 Income distribution Middle 40 Bottom 40 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% decline >=30% decline < 30% same increase <30% increase>=30% Moreover, a low rate of savings make it more difficult for households to cushion the shock 58% households reported no savings to deal with any reduction in income. The situation is more difficult for poorer households, and households with no member working in February 70% 64% 58% 55% 50% 36% Bottom 40 Middle 40 Top 20 Not working Working Income distribution Employment status in Feb All Even among household with savings, 76% of them would run out of savings in the next 6 months. Poor households and households with older household head are more likely to run out savings 100 90 87 76 75 78 80 70 70 62 60 50 41 40 31 30 24 20 10 0 Bottom 40 Middle 40 Top 20 Male Female Age 25-49 Age 50-64 Age 65+ All Income Gender of household Age of household head head SUBJECTIVE WELLBEING It is expected that many households faced difficulty to make ends meet during the pandemic 50% of all households reported difficulties with their current income level to pay for their needs in June. Poor households suffered the most. So did rural households, or households with older household head or with household head not working prior to the crisis. • 70% of poorer households report they’re finding it difficult making ends meet. • About one-third of the traditionally “middle class” households report difficulties, even though these are traditionally less economically vulnerable. • However, female-headed households did not appear to face more difficulties than male-headed households. • The share of those having difficulty accessing food and sanitation as normal is likely a factor of logistics during lockdown as well as financial situation deterioration. Total Access to sanitation Top 20 Middle 40 Bottom 40 Total Access to food Top 20 Middle 40 Bottom 40 Total Making ends Top 20 meet Middle 40 Bottom 40 0% 10% 20% 30% 40% 50% 60% 70% 80% % households Great difficulty Difficulty Compared to February (pre-crisis), a quarter of households reported deteriorating economic situation The pattern is similar across population groups (gender, age, location, employment status) • Poorer households report greater difficulty making ends meet in June compared to February. • But the difficulty gaps by household heads’ gender, age, location, and employment status is not significant. More difficult to make ends meet in June compared to February 33% 26% 21% 17% Bottom 40 Middle 40 Top 20 Total The level of economic stress reported due to the COVID-19 pandemic is high across the board 81% households indicated that they were worried about the economic well-being of themselves and their family members. 20% Strongly agree 61% Agree 14% Disagree 3% Strongly disagree But economic stress varies across the population Women, young, and poor households had significantly higher stress 100 90 % household worried about economic wellbeing 80 70 60 61 61 61 63 50 59 60 57 57 57 55 40 30 20 23 26 10 20 18 19 21 21 21 17 15 0 Male Female Age 20-29 Age 30-39 Age 40-49 Age 50-64 Bottom 40 Middle 40 Top 20 All Gender Age Income level Strongly agree Agree Moreover, women had more stress in multiple dimensions 47% women feeling nervous about the current situation, and 32% being stressful about leaving the house, compared to 43% and 27% for men respectively 50 47 45 45 43 40 35 32 % household agree 30 30 27 25 20 15 10 5 0 Nervous about the current situation Stress about leaving the house All Male Female 92% women worrying about their health or health of their family members, compared to 88% men 93 92 92 % households agree 91 90 90 89 88 88 87 86 All Male Female While 90% households worrying about their health situation, access to health services is relatively high On average, 40% households needed medical care during the pandemic. The rate is higher among women, lower income, and older people Age 50-64 45 Age 40-49 39 Age Age 30-39 39 Age 20-29 27 Top 20 28 Income Middle 40 42 Bottom 40 41 All Gender Female 42 Male 38 40 0 10 20 30 40 50 % households need medical care Among those in need of healthcare, 84% received treatment. The difference is not significant between men and women. However, the rate is significantly lower among young and richer people. This observation may be explained by the differences in types of medical treatment needed for different population groups Age 50-64 85 Age 40-49 82 Age Age 30-39 86 Age 20-29 76 Top 20 68 Income Middle 40 85 Bottom 40 85 All Gender Female 82 Male 86 84 0 20 40 60 80 100 % households received medical care EDUCATION While school enrollment is high across the board… 100 99 97 97 97 97 97 97 97 96 95 95 90 85 80 % children 75 70 65 60 55 50 Male Female Rural Urban Bottom 40 Middle 40 Top 20 Working Not working All Gender of child Location Income distribution At least one adult working and schools quickly offered some types of distant learning (online, printed materials, distant homework, etc.) during the pandemic… Any type of distant learning 100 100 99 99 99 98 98 98 97 97 96 95 90 85 80 % children 75 70 65 60 55 50 Male Female Rural Urban Bottom 40 Middle 40 Top 20 Working Not working All Gender of child Location Income distribution At least one adult working But access to online learning — the most advance and interactive distant learning — varies significantly across demographic groups Girls, rural students, students from lower income households, and students from households with no adults working had limited opportunities. Online distant learning 100 98 91 92 93 92 88 88 90 90 84 82 84 79 80 70 66 60 % children 50 40 30 20 10 0 Nursery Primary High Male Female Rural Urban Bottom Middle Top 20 Working Not school 40 40 working All School Gender of child Location Income distribution At least one adult working In addition, distribution of distant learning equipment from school was uneven across groups And the level is relatively low — only one in three students received a tablet from school • Primary schools and urban schools were more likely to give out tablet. • Interestingly, the poorest and the richest students had equal chance of receiving tablets from school, which may reflect the fact that richer schools were more equipped, and other schools targeted poor students well. However, students in middle-class families were less likely to receive support. Tablet from school 50 45 44 40 39 35 35 32 33 31 30 30 28 25 22 20 15 10 7 5 0 Primary High Male Female Bottom 40 Middle 40 Top 20 Urban Rural school All School Gender of child Household income Location However, to support distant learning, almost all students have access to tablets with internet The level of shared tablets within a households is higher among poor and rural households. There is no significant difference by children’s gender or school levels 100% 13 90% 18 27 26 20 26 26 27 33 80% 36 70% 60% 50% 87 40% 78 81 73 72 73 71 74 67 30% 63 20% 10% 0% Primary High Male Female Bottom Middle 40 Top 20 Urban Rural school 40 All School Gender of child Household income Location Have individual tablet Have shared tablet Lots of support from households were provided to cope with distant learning 84 percent of students received help from another household members, often the non-working members • Younger students are more likely to require support. 93 percent of primary student received help from home to do their distant learning schoolwork. • Richer students are also much likely to receive support, partially because they could afford more than poorer students. Almost all students in the top 20 percent of the income distribution got private help. Help with schoolwork 100 96 93 90 88 84 84 83 85 82 79 80 70 67 60 50 40 30 20 10 0 Primary High Male Female Bottom 40 Middle 40 Top 20 Urban Rural school All School Gender of child Household income Location On average, students spent 4.7 hours per day on schoolwork The amount of time is slightly higher among girls, urban students, and better-off students. This could partially be explained by the fact that urban and better-off students were more likely to have individual tablets and have help with homework. Average number of hours 6 5.4 5.2 5.1 5 5 4.8 4.7 4.6 4.4 4.3 4.3 4 3 2 1 0 Primary High Male Female Bottom 40 Middle 40 Top 20 Urban Rural school All School Gender of child Household income Location But distant learning took a toll on student’s mental health One in four students were either unhappy or experienced problems with it. Boys, younger students, students from households with no adults working, and especially students from higher income households (despite all the support), were more likely to have trouble coping with distant learning. Have trouble or unhappy 35 33 32 30 30 29 27 25 24 % children having distant learning 22 20 17 15 15 14 10 5 0 Primary High Male Female Bottom 40 Middle 40 Top 20 Working Not school working All School Gender of child Income At least one adult working Parents were slightly worried about school reopening, especially poor parents or parents with high school students The most concern is about the child getting infected 23 percent parents with high school students worried that their children may get affected, much higher than parents with young children 25 23 20 19 % parents worrying... 17 16 15 15 14 14 12 12 12 11 10 10 5 0 Child may get infected Child may infect other School won't have Monitoring over safety HH members safety measure will be weak All Primary school High school 19 percent of parents in the lowest income bracket worried about their child’s infection. The rate is about half among parents with higher income 20 19 18 17 16 14 14 13 12 12 11 11 11 % parents 12 10 9 9 8 6 4 2 0 0 Child may get infected Child may infect other School won't have Monitoring over safety HH members safety measure will be weak Bottom 40 Middle 40 Top 20 Access to healthcare services for children is quite high during the pandemic, although children in rural areas tend to have slightly less access About half of the children need health services during the pandemic. The demand is lower among older children, and among higher income families income Location Rural 45 Urban 43 Gender Household Top 20 30 Middle 40 50 Bottom 40 43 All School of child Female 43 Male 45 High school 35 Primary 48 44 0 10 20 30 40 50 60 % children need healthcare Among those in need, nearly 90 percent received heath care services. A lower rate was reported in rural areas which could be due to the availability of health staff and facility. Meanwhile, a lower rate among better off households could reflect the type of health care required, or the urgency of treatment income Location Rural 87 Urban 93 Gender Household Top 20 80 Middle 40 90 Bottom 40 88 All School of child Female 91 Male 87 High school 86 Primary 90 89 70 75 80 85 90 95 % children receiving health care