4. The impact of the crisis on households in Greece
The impact of the crisis on households in Greece
How have the implications of the economic crisis and the related austerity measures impacted on the lives of Greek residents? This chapter takes a look at the experience of Greek households in coping with recent challenges, and how the crisis has affected their material and subjective well-being and their trust in political institutions.
With more than one in four individuals of working age unemployed and six consecutive years of negative annual economic growth from 2008 onwards, the economic crisis experienced in Greece has been deep and long-lasting. By the end of 2015, GDP had contracted by 26 per cent compared to 2008 levels;1 over the same period, total unemployment went up by 17.1 percentage points and youth unemployment increased by almost 28 percentage points to 49.8 per cent (see Chart 4.1). By comparison, total unemployment and youth unemployment increased by 3.3 and 6.3 percentage points, respectively, in the euro area between 2008 and 2015.
How have the implications of the economic crisis and the related austerity measures impacted on the lives of Greek residents? This chapter takes a look at the experience of households in the country, examines their perceptions and sheds light on the transmission mechanisms and the coping strategies they adopted to overcome the challenges posed by economic difficulties. It then investigates how the crisis has affected their material and subjective well-being and their trust in political institutions.
This chapter uses novel data from 1,500 households collected as part of the third round of the Life in Transition Survey (LiTS III), which features a dedicated module on the “Impact of the crisis and austerity”. This module is composed of two sections. The first one is largely based on the “Impact of the crisis” module implemented in 2010 during the second round of LiTS (LITS II), and elicits perception-based and factual information on how the crisis affected households, which mechanisms they used to cope, and whether they had savings in 2010 which they could use to cover daily expenses and higher taxes during the austerity measures. This new information can be compared to LiTS II data to uncover differences between the Greek experience and that of over 38,000 households in 35 different countries during the period 2008-10. The second section leverages questions on the respondents’ political preferences, including their voting behaviour in the most recent national elections and in the July 2015 referendum for the approval of the bailout conditions proposed by the European Commission, the European Central Bank and the International Monetary Fund.
SOURCE: Eurostat (2016).
NOTE: Total unemployment is computed as a percentage of the total labour force that is without work but seeking employment, while youth unemployment refers to the percentage of the labour force aged between 15 and 24 years without work but seeking it.
LiTS III data can be used to estimate the impact of the crisis on people in both a subjective and an objective way. The subjective measure is obtained using responses to a perception-based question that reads: “As you know, an economic crisis is affecting Greece. How much, if at all, has this crisis affected your household in the past five years?”. Affected households are defined as the ones who answered either “a fair amount” or “a lot” as opposed to “a little” or “not at all”. The objective measure instead represents the percentage of households that experienced at least one of the following negative income shocks over the reference period: the family business was closed, one of the family members lost their job, saw their wages, pensions or working hours reduced, or had their wages delayed or suspended.
Chart 4.2 shows the percentage of households that were affected by the crisis in Greece in the five years prior to the survey according to the subjective and the objective measures. It also compares the Greek experience with that of other countries in 2008-10. The results are striking: over 92 per cent of Greek respondents believe that the crisis has affected their households “a fair amount” or “a lot”, while 76 per cent of them experienced a negative income shock. In contrast, one in two households in the transition region and about one in three in the western European comparator countries reported having been affected by the crisis and having experienced a negative income shock in the two years prior to 2010. Not only do these results indicate that the Greek crisis affected relatively more households than the Great Recession, but they also show that a large proportion of Greek respondents perceive themselves as affected, even in absence of negative income shocks.
While Chart 4.2 shows that the crisis was very pervasive and widespread, econometric tests reported in Table 4.1 indicate that families where the household head has either no or only a limited education or where the household head is a woman were disproportionately affected.2 The impact of the crisis appears to have been less noticeable for families with a young household head (that is, one aged 18-24). A closer look at the determinants of the objective measure does indeed suggest that a larger number of families with an older household head experienced different types of negative income shocks compared to younger families. The highest incidence of job losses and delay or suspension of wages was reported by the group of families with a household head aged 25 to 39 years, whereas reductions in wages or pensions were more prevalent among households with a head older than 65.
Bigger families, ethnic minorities and households living in Crete as opposed to mainland Greece also report having been more affected by the crisis than their counterparts.3 The impact of the crisis in Crete is likely to have been more severe due to the higher incidence of delayed or suspended wages and reduction in the number of working hours compared to the rest of the country.4 Lastly, income proxies are, predictably, negatively correlated with crisis impact,5 while asset ownership does not appear to matter.6 Once proxies for income are controlled for, the coefficient for the gender of the household head loses significance, most likely indicating that female-headed households are more likely to report having been affected by the crisis due to their being worse off than male-headed ones.
SOURCE: LiTS II (2010), LiTS III (2016) and author’s calculations.
NOTE: The figures for Greece refer to the 2010-16 period and were calculated using LiTS III data. The figures for the remaining countries refer to the 2008-10 period and were calculated using LiTS II data.
| (1) | (2) | (3) | (4) | (5) | |
|---|---|---|---|---|---|
| Affected by the crisis (subjective measure, four-point scale) | |||||
| Male household head | -0.195** | -0.201** | -0.191** | -0.179* | -0.142 |
| (0.087) | (0.092) | (0.092) | (0.095) | (0.104) | |
| Household head is 18-24 years old | -0.381 | -0.554** | -0.496** | -0.527** | -0.415* |
| (0.245) | (0.218) | (0.240) | (0.243) | (0.245) | |
| Household head is 40-64 years old | 0.212** | 0.117 | 0.130 | 0.122 | 0.191** |
| (0.090) | (0.087) | (0.087) | (0.088) | (0.095) | |
| Household head is 65 years old or above | 0.328*** | 0.049 | 0.039 | -0.010 | 0.163 |
| (0.119) | (0.115) | (0.121) | (0.127) | (0.126) | |
| Ethnic minority | 0.438** | 0.365** | 0.360** | 0.319* | 0.235 |
| (0.177) | (0.174) | (0.176) | (0.172) | (0.171) | |
| Household size (total) | 0.088*** | 0.089*** | 0.081** | 0.102*** | 0.077** |
| (0.031) | (0.032) | (0.032) | (0.032) | (0.033) | |
| Mainland Greece | -0.357** | -0.331** | -0.303** | -0.312* | -0.418** |
| (0.170) | (0.157) | (0.154) | (0.159) | (0.176) | |
| Household head completed secondary education | -0.332*** | -0.332*** | -0.291** | -0.166 | |
| (0.121) | (0.121) | (0.117) | (0.125) | ||
| Household head completed tertiary education | -0.730*** | -0.737*** | -0.678*** | -0.380** | |
| (0.136) | (0.138) | (0.137) | (0.149) | ||
| Household can afford a one-week holiday each year and meat, chicken or fish every second day | -0.657*** | ||||
| (0.094) | |||||
| Household can meet unexpected expenditures | -0.341*** | ||||
| (0.114) | |||||
| Observations | 1431 | 1465 | 1431 | 1431 | 1431 |
| Pseudo R2 | 0.017 | 0.038 | 0.037 | 0.038 | 0.095 |
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: This table reports the results of ordered probit regressions. Standard errors in parentheses are clustered at the PSU level. *, **, and *** denote values that are statistically significant at the 10, 5 and 1 per cent levels, respectively. The reference categories for the age and education of the household head are “Household head is 25-39 years old” and “Household head completed either no education or primary education”. “Household size (total)” is the total number of members currently living in the household. The specifications reported in column 1 and in columns 3-5 include an indicator signalling whether or not the household had savings prior to 2010. The effect of the indicator is not statistically significant. The specifications in columns 4-5 also include three controls for whether the household owns a car, a computer and a washing machine. None of these is statistically significant.
The crisis affected Greek families in different ways, ranging from income reductions to job losses and business closures. Chart 4.3 compares the transmission channels of the 2008-09 crisis in south-eastern Europe (SEE), the transition region and five western European comparator countries with the more recent Greek experience. According to the data, over the period 2010-16 families were impacted mainly through wages or pensions cuts and job losses: almost 44 per cent of Greek households saw their wages or pensions reduced and in more than 24 per cent of the interviewed families at least one member lost their job. While these figures are comparable to the averages for SEE and other transition countries, the incidences of wages or pension reductions and job losses were much lower in the comparator countries, by about 20 and 10 percentage points respectively. Greek respondents that suffered a reduction in their wages or pensions registered, on average, a 24 per cent cut with respect to the original amount.7
Households in western European comparator countries were slightly more likely to adjust the number of their working hours downwards compared to Greek ones, perhaps reflecting a higher degree of contract flexibility that might have reduced the need for employers to decrease wages or eliminate jobs. On the bright side, households in Greece seem to have experienced a lower incidence of delayed or suspended wages and fewer closures of family businesses than those in SEE and the transition region, with the latter result being more in line with the corresponding average for the comparator countries.
Overall, and predictably, the Greek households that were most affected by the crisis report a higher incidence of job losses for at least one family member and reductions in the number of working hours (see Chart 4.4). Differences in the extent to which affected and unaffected families were impacted by wages or pensions reductions and wages delay or suspensions are not statistically significant. Wages or pension reductions were instead more frequent in female-headed households, while a larger fraction of male-headed households experienced shorter working hours or delays or suspensions of wages. Male-headed and female-headed households were affected by the remaining negative income shocks in a similar way, as the differences in the reported figures are not statistically significant.
SOURCE: LiTS II (2010), LiTS III (2016) and author’s calculations.
NOTE: The chart shows the percentage of households that experienced any of the reported shocks as the result of the economic crisis in Greece and in other countries in the five and two years prior to the survey, respectively. The averages include all the households where the respondent reports having been affected by the crisis “a little”, “a fair amount” or “a lot”. The category “job losses” includes all respondents who stated that the household head or another member of the household lost their job. The figures for Greece refer to the 2010-16 period and were calculated using LiTS III data. The figures for the remaining countries refer to the 2008-10 period and were calculated using LiTS II data. For the purpose of this chart, SEE includes the following countries: Albania, Bosnia and Herzegovina, Bulgaria, FYR Macedonia, Kosovo, Montenegro, Romania and Serbia. Comparator countries include France, Germany, Italy, Sweden and the United Kingdom.
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: The chart shows the percentage of Greek households that experienced any of the reported shocks as the result of the economic crisis in the five years prior to the survey. The averages broken down by gender of the household head include all the households where the respondent reports having been affected by the crisis “a little”, “a fair amount” or “a lot”. For a description of the category defined as “job losses”, refer to the notes of Chart 4.3.
Almost
44%
of Greek households saw their wages or pensions reduced during 2010-16.
Greek households had to resort to a variety of strategies to cope with the consequences of the crisis. These included passive strategies, such as reducing the consumption of specific goods, discontinuing subscriptions to services (such as television, phone and internet) or postponing payments, and active strategies, such as obtaining an additional job or increasing the number of working hours.8 According to LiTS III data, over 94 per cent and over 51 per cent of affected households had to reduce the consumption of non-necessities and necessities, respectively (see Chart 4.5).9 ,10 In comparison, only 64 and 19 per cent of families reduced their consumption of non-necessities and necessities in the western European comparator countries in the period 2008-10. These results are particularly alarming in that necessities comprise basic, essential items such as staple foods, visits to the doctor when ill and regular medications. They also show how deep and dramatic the Greek crisis has been for the average household, and how Greeks have suffered considerably more during this time than residents of other countries did during the 2008-09 crisis.
Looking at spending categories separately, the biggest cuts were registered for luxury goods and vacations (75 per cent of the respondents state they had to reduce their consumption of both), followed by staple foods such as milk, fruit, vegetables or bread and use of own car (both at 47 per cent), alcohol (46 per cent) and delayed payment of utilities (45 per cent) (see Chart 4.6).
While the consumption of non-necessities decreased across the board, those Greek households that were most affected by the crisis also had to disproportionately reduce their consumption of necessities, delay payments on utilities or cut their television, phone or internet services, delay or default on loan or mortgage payments or re-negotiate their tax schedule in order to keep up with payments. Chart 4.7 compares the coping strategies of severely affected and less affected households based on the perception-based measure of crisis impact; these discrepancies are also confirmed when the objective measure is used instead. Households that managed to save up some money before the austerity measures were introduced appear to have been less financially constrained than their counterparts and consequently had to cut back less on consumption.11
Additional analysis shows that almost 62 per cent of female-headed households reduced their consumption of necessities, whereas 47 per cent of male-headed families resorted to the same coping strategy. Conversely, 96 per cent of male-headed households cut their consumption of non-necessities, while the corresponding figure for female-headed ones is 89 per cent. Lastly, families where the head completed at least secondary education saw a lower frequency of reduction in their consumption of necessities in comparison to families where the head had no education or only primary education (47 per cent as opposed to 65 per cent), but a higher incidence of reduction in their consumption of non-necessities (97 per cent as opposed to 84 per cent).
Columns 1 to 3 of Table 4.2 provide econometric evidence that male-headed households, those with a highly educated household head, with accumulated savings or with a higher income (as proxied by two indicators) had to reduce their consumption of necessities less often, even after controlling for a selection of individual and household characteristics. In contrast, ethnic minorities were more likely to cut their consumption of necessities as a coping strategy. Looking at columns 4 to 6, it appears that households where the head is a man or where the head is highly educated were more likely to cut their consumption of non-necessities than their peers; this is possibly due to higher consumption levels in the pre-crisis period. Taken together, these results show that female-headed families and those with a lower level of education were disproportionately affected by the crisis and had to make bigger sacrifices than their peers, even after taking their size, residence and wealth into consideration.
In addition, Chart 4.5 shows that, compared to Greece, households in SEE and in the transition and comparator countries were more likely to take proactive steps to counter the effect of the crisis. These included increasing the number of working hours, having a family member work two jobs or encouraging a family member who was previously not working to take up employment. This stark difference in behaviour is likely attributable to the sharp increase in the unemployment rate in Greece as compared to the other countries following the 2008-09 crisis. Between 2008 and 2010 the change in unemployment rates ranged from 1.7 percentage points in Italy to 2.4 percentage points in Sweden; in Germany, unemployment decreased from 7.4 to 7 per cent over the same period. By comparison, unemployment in Greece rose by 12.2 percentage points, from 12.7 per cent in 2010 to 24.9 per cent in 2015.12 This has possibly limited the availability of options and work opportunities, even for those Greek individuals who might have wanted to start a job or work longer hours.
Households in Greece who managed to increase their working hours or get an additional job are more likely to state that the crisis has affected them only “a little”. This is consistent with the idea that families that adopted more proactive behaviours were more capable of coping with the adverse consequences of the crisis. Interestingly, a higher percentage of male-headed households managed to counteract the effects of the crisis by increasing their working hours or getting an additional job (4.8 per cent and 2 per cent respectively, in contrast to only 2.6 and 0.7 per cent of female-headed households). This finding is confirmed by econometric tests, shown in the last three columns of Table 4.2. Even controlling for a wide selection of individual and household characteristics, including the number of adult members, male-headed households were more likely to be able to respond to the challenges posed by the crisis in a proactive way. Ethnic minorities, by contrast, were less likely to do so.
SOURCE: LiTS II (2010), LiTS III (2016) and author’s calculations.
NOTE: The chart shows the percentage of households that took any of the reported measures as the result of a decline in income or other economic difficulty in Greece and in other countries in the five and two years prior to the survey, respectively. The averages include all the households where the respondent reports having been affected by the crisis “a little”, “a fair amount” or “a lot”. For a description of the categories defined as “non-necessities” and “necessities”, refer to footnotes 9 and 10. The category “got additional job” includes all respondents who declared that someone in the household who was already working took a second job or additional work, or that someone who was not working before found a new job (either part-time or full-time). The figures for Greece refer to the 2010-16 period and were calculated using LiTS III data. The figures for the remaining countries refer to the 2008-10 period and were calculated using LiTS II data. For the purpose of this chart, SEE includes the following countries: Albania, Bosnia and Herzegovina, Bulgaria, FYR Macedonia, Kosovo, Montenegro, Romania and Serbia. Comparator countries include France, Germany, Italy, Sweden and the United Kingdom.
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: The chart shows the percentage of households that took any of the reported measures as the result of a decline in income or other economic difficulty in Greece in the five years prior to the survey. The averages include all the households where the respondent reports having been affected by the crisis “a little”, “a fair amount” or “a lot”.
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: The chart shows the percentage of households that took any of the reported measures as the result of a decline in income or other economic difficulty in Greece in the five years prior to the survey. The averages broken down by availability of savings include all the households where the respondent reports having been affected by the crisis “a little”, “a fair amount” or “a lot”. For a description of the categories defined as “non-necessities” and “necessities”, refer to footnotes 9 and 10. For a description of the category defined as “got additional job”, refer to the notes of Chart 4.5. “Some savings” refers to those households which had savings in 2010 that could be used to cover daily expenses and higher taxes during the period of austerity starting in 2010.
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | |
|---|---|---|---|---|---|---|---|---|---|
| Household coped with crisis by cutting expenses on necessities | Household coped with crisis by cutting expenses on non-necessities | Household coped with crisis by finding additional job or increasing hours | |||||||
| Male household head | -0.323*** | -0.271*** | -0.198** | 0.295** | 0.177 | 0.119 | 0.331** | 0.351** | 0.332** |
| (0.095) | (0.091) | (0.100) | (0.117) | (0.120) | (0.140) | (0.149) | (0.154) | (0.147) | |
| Household head is 18-24 years old | -0.458* | -0.555* | -0.736*** | -0.244 | 0.041 | 0.114 | -0.392 | -0.414 | -0.412 |
| (0.267) | (0.283) | (0.266) | (0.317) | (0.486) | (0.509) | (0.325) | (0.330) | (0.328) | |
| Household head is 40-64 years old | -0.076 | -0.018 | 0.027 | 0.283 | 0.171 | 0.242 | -0.099 | -0.078 | -0.073 |
| (0.123) | (0.124) | (0.123) | (0.230) | (0.191) | (0.184) | (0.150) | (0.151) | (0.153) | |
| Household head is 65 years old or above | -0.071 | -0.008 | 0.030 | -0.105 | -0.196 | -0.045 | -1.210*** | -1.174*** | -1.110*** |
| (0.151) | (0.158) | (0.178) | (0.272) | (0.256) | (0.268) | (0.288) | (0.295) | (0.298) | |
| Ethnic minority | 0.603*** | 0.519*** | 0.248 | -0.429 | -0.321 | -0.172 | -0.753*** | -0.767*** | -0.695** |
| (0.193) | (0.195) | (0.198) | (0.283) | (0.273) | (0.263) | (0.283) | (0.290) | (0.287) | |
| Household size (total) | -0.051 | -0.051 | -0.006 | 0.162* | 0.205** | 0.134 | |||
| (0.038) | (0.038) | (0.040) | (0.087) | (0.094) | (0.089) | ||||
| Household size (adults) | -0.109 | -0.099 | -0.128 | ||||||
| (0.082) | (0.084) | (0.088) | |||||||
| Mainland Greece | 0.091 | 0.070 | -0.103 | -0.320 | -0.261 | -0.296 | 0.087 | 0.083 | 0.110 |
| (0.179) | (0.188) | (0.194) | (0.416) | (0.409) | (0.397) | (0.290) | (0.295) | (0.301) | |
| Household head completed secondary education | -0.311** | -0.285** | -0.051 | 0.749*** | 0.760*** | 0.659*** | 0.129 | 0.145 | 0.048 |
| (0.130) | (0.134) | (0.133) | (0.210) | (0.225) | (0.206) | (0.244) | (0.244) | (0.252) | |
| Household head completed tertiary education | -0.766*** | -0.716*** | -0.264 | 0.864*** | 0.831*** | 0.626*** | 0.052 | 0.094 | -0.077 |
| (0.163) | (0.172) | (0.181) | (0.217) | (0.236) | (0.211) | (0.229) | (0.232) | (0.253) | |
| Household can afford a one-week holiday each year and meat, chicken or fish every second day | -0.640*** | 0.283* | 0.128 | ||||||
| (0.117) | (0.157) | (0.176) | |||||||
| Household can meet unexpected expenditures | -0.446*** | -0.171 | 0.068 | ||||||
| (0.108) | (0.203) | (0.156) | |||||||
| Observations | 1402 | 1369 | 1369 | 1402 | 1369 | 1369 | 1397 | 1363 | 1363 |
| Pseudo R2 | 0.053 | 0.056 | 0.143 | 0.170 | 0.175 | 0.190 | 0.088 | 0.092 | 0.098 |
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: This table reports the results of probit regressions. Standard errors in parentheses are clustered at the PSU level. *, **, and *** denote values that are statistically significant at the 10, 5, and 1 per cent levels, respectively. The reference categories for the age and education of the household head are “Household head is 25-39 years old” and “Household head completed either no education or primary education”. “Household size (total)” is the total number of members currently living in the household. “Household size (adults)” is the total number of members aged 18 or above currently living in the household. The specifications reported in columns 2-3, 5-6 and 8-9 include an indicator signalling whether or not the household had savings prior to 2010. The effect of the indicator is not statistically significant, except in column 2, where it is negative and significant. The specifications in columns 3, 6 and 9 also include three controls for whether the household owns a car, a computer and a washing machine. Only the first one in column 3 is negative and significant; none of the others is statistically significant.
About
24%
of Greek respondents are currently satisfied with their life.
The results of the previous sections clearly show that the protracted crisis has taken a severe toll on the well-being of Greek residents. Further analysis of the data also indicates that, in 2016, Greek households cannot afford the same level of material well-being as their counterparts in other countries. Chart 4.8 presents data reflecting the current situation in Greece and the post-crisis situation in other countries by means of three different indicators. The first one signals whether or not a household can afford a one-week holiday every year (including a stay at a second home, country house or at friends or relatives). The second indicator shows whether a household can afford to consume meat, chicken or fish every second day. The last one indicates whether the family can meet unexpected expenditures of an amount equal to the domestic poverty line with its own resources.
Data suggest that Greeks have a level of material well-being similar to the one seen in SEE and the transition region, but the country fares worse than Germany and Italy according to all indicators. Only 41 per cent of Greek households can afford to have a holiday each year, as opposed to almost 79 per cent in Germany and over 61 per cent in Italy. In addition, about 58 per cent of Greek households can afford to consume meat, chicken or fish every second day, while virtually every family can do so in Germany. In Italy, 8 out of 10 households report being able to afford meat, chicken or fish this often.
The majority of Greek respondents state they cannot meet unexpected expenditures, even though this varies significantly by the magnitude of the impact of the crisis: while just over 43 per cent of affected households can afford unexpected expenditures, almost 72 per cent of the unaffected families can. Interestingly, by contrasting the level of material well-being of households that reported having being affected by the crisis “a fair amount” or “a lot” with those that were either affected “a little” or not affected at all, it appears that the latter have a level of material well-being at least comparable to, if not higher than, Germany and Italy. Other data also show that only 71 per cent of Greek households can afford adequate heating for their homes, while over 86 per cent and 98 per cent of households in transition and comparator countries can. The corresponding figures for the affected and unaffected Greek households are 69 and 94 per cent respectively.
Given the discrepancy in material well-being between Greek residents and other respondents, it is perhaps not surprising that subjective well-being levels are lower among the Greeks than in the rest of the sample. Virtually every interviewed respondent believes that the economic and the political situations did not improve over the four years prior to the survey (see Chart 4.9).13 More than nine in ten Greeks affected by the crisis are not happy with their financial situation, and only about one in four is satisfied with his or her life.14 Happiness levels among those unaffected are in these instances similar to the averages for Germany and Italy.
Are Greek residents generally dissatisfied with their country’s economic and political situation, and unhappy about their financial situation and their life, or are these differences driven by individual characteristics? Results from a simple econometric model of satisfaction, which controls for an array of individual and household characteristics as well as country fixed effects, show that Greek people are indeed less satisfied and happy than their counterparts in other countries.15 Additional econometric tests presented in Table 4.3, which are run on the Greek data only and control for the subjective measure of crisis perception, also indicate that the households that were most affected by the crisis systematically report lower levels of life satisfaction than their peers and thus corroborate the findings of Chart 4.9 (particularly the discrepancy between the grey and the violet diamonds).16
Considering the general dissatisfaction with the economic situation, one would expect the economy to be the main concern of the Greek people. Indeed, when asked which problems they consider most pressing and deserving of the government’s attention, the Greeks list the economy and unemployment at the top, followed by health (see Chart 4.10). Cumulatively, 74 per cent of respondents are concerned with the first two problems, as opposed to about 45 per cent of the population in the transition region and 11 and 46 per cent in Germany and Italy respectively. Despite the number of refugees that have reached Greece since the summer of 2015, only 4 per cent of interviewed Greek respondents believe that immigration is the most important issue to be addressed; however, a more substantial 36 per cent of Greek respondents place it within the top three issues they feel their government should focus on. By contrast, over 46 per cent of German respondents and 16 per cent of Italian respondents think that immigration is the most important problem in their respective countries, with 69 per cent of Germans and 49 per cent of Italians placing it within the top three.
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: The chart shows the percentage of households that can afford the consumption of the reported items. The last category includes households that can afford unexpected expenditures “easily” and “with difficulty”. The diamonds indicate the variation in results by magnitude of the impact of the crisis. The grey diamonds show the percentage of unaffected respondents who can afford the reported items, while the violet diamonds show the percentage of affected respondents who can afford the same items. For the purpose of this and the following charts, SEE includes the following countries: Albania, Bosnia and Herzegovina, Bulgaria, Cyprus, FYR Macedonia, Greece, Kosovo, Montenegro, Romania and Serbia.
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: The chart shows the percentage of respondents who either “agree” or “strongly agree” with the statements reported on the horizontal axis. The diamonds indicate the difference in the results by magnitude of the impact of the crisis. The grey diamonds show the percentage of unaffected respondents who either agree or strongly agree with the reported statements, while the violet diamonds show the percentage of affected respondents who either agree or strongly agree with them. For a list of the countries included in SEE, refer to the notes of Chart 4.8.
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: The chart shows the percentage of respondents that list any of the reported issues as the most important problem that the government should address. For a list of the countries included in SEE, refer to the notes of Chart 4.8.
| (1) | (2) | (3) | (4) | |
|---|---|---|---|---|
| The economic situation in our country is better today than around 4 years ago | The political situation in our country is better today than around 4 years ago | All things considered, I am satisfied with my financial situation as a whole | All things considered, I am satisfied with my life now | |
| Affected by the crisis a fair amount or a lot | -0.479*** | -0.344** | -0.996*** | -0.593*** |
| (0.133) | (0.154) | (0.138) | (0.116) | |
| Observations | 1500 | 1496 | 1500 | 1500 |
| Pseudo R2 | 0.022 | 0.013 | 0.116 | 0.077 |
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: This table reports the results of ordered probit regressions. Standard errors in parentheses are clustered at the PSU level. *, **, and *** denote values that are statistically significant at the 10, 5, and 1 per cent levels, respectively. All the specifications include the following controls: gender of the respondent; age of the respondent; whether the respondent belongs to an ethnic minority; whether the respondent lives in mainland Greece; education of the respondent; whether the household of the respondent owns a car, a computer and a washing machine; whether the household can afford a one-week holiday each year and meat, chicken or fish every second day; and whether the household can meet unexpected expenditures. In column 4, four additional controls are included: the total number of household members currently living in the household; whether the respondent is married; whether the respondent is divorced or separated; and whether the respondent is a widow or widower.
The low level of satisfaction with the political situation in Greece goes hand-in-hand with a low opinion of the work of the national government and a low level of trust in the country’s political institutions. When asked to rate the performance of the incumbent national government, only 11 per cent of Greek respondents describe it as “good” or “very good”, in contrast to 25 per cent of the population in transition countries and Germany (see Chart 4.11). In this regard, Italy fares worse than Greece, with only 8 per cent of its residents expressing a positive opinion of their national government.
While recognition for the work of regional governments is more in line with results from other countries (but is still rather low), appreciation for the performance of the local governments in Greece is higher than the averages for the transition region and both comparator countries. Moreover, it is noteworthy that a higher proportion of Greek respondents assign a “good” or “very good” rating to local governments than national ones. Local governments are usually associated with an individual (such as the mayor) and are therefore seen as more personable, which can positively influence the perceptions of people to a certain extent, especially in smaller communities where this person is visible. In addition, Greek respondents might be more satisfied with or have more trust in their local institutions if they feel that there are more opportunities for their opinions or discontent to be heard than there are at the national level.17
Examining trust in institutions delivers similar results (see Chart 4.12). Trust in local and regional governments in Greece is higher than trust in the national government, but now on average lower compared to that of other countries. Trust in the President, the national government and the parliament is also much lower than in SEE, the transition region and Germany and Italy, at 24, 11 and 8 per cent respectively, while only 3 per cent of the Greek population deem the country’s political parties to be trustworthy. Additional analysis shows that Greeks fare worse than the other countries in LiTS III when it comes to trusting banks, foreign investors and society as a whole.18 Interestingly, trust levels are lower among respondents affected by the crisis and the differences are statistically significant for all the institutions reported in Chart 4.12, with the sole exception of political parties.19
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: The chart shows the percentage of respondents who rate the performance of the institutions reported on the horizontal axis as “good” or “very good”. For a list of the countries included in SEE, refer to the notes of Chart 4.8.
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: The chart shows the percentage of respondents who believe that the institutions reported on the horizontal axis deserve “some trust” or “complete trust”. For a list of the countries included in SEE, refer to the notes of Chart 4.8.
The majority of Greek respondents blame political parties for the economic crisis (see Chart 4.13). When asked who they believe is responsible for the current situation, with the option to give more than one answer, around 70 per cent of respondents mention the centre-left political party PASOK and centre-right political party New Democracy. These two parties dominated the political scene and took turns at leading the national government from the restoration of democracy in 1974 until the elections in January 2015, which culminated in the Coalition of the Radical Left (SYRIZA) winning and assuming power. These responses can explain PASOK’s electoral decline (from 43.9 per cent in the parliamentary elections of 2009 to 4.7 per cent and 6.3 per cent in the parliamentary elections of January and September 2015 respectively) which paved the way for SYRIZA’s rise from 4.6 per cent in 2009 to 36.3 per cent and 35.5 per cent in the elections of 2015.
The European Union (EU), the International Monetary Fund (IMF) and the European Central Bank (ECB) are the runners-up, with between 42 and 50 per cent of respondents believing those organisations are responsible for the crisis. This is probably because the EU’s executive body, the European Commission, the IMF and the ECB formed a tripartite committee, known as the “Troika”, which negotiated the terms of Greece’s bailout agreements.
An analysis of the voting behaviour of Greek respondents shows a correlation between voting preferences in the July 2015 referendum on the approval or rejection of bailout conditions and the institutions that respondents see as responsible for the crisis (Table 4.4). People who believe the EU is to blame were less likely to accept the proposed bailout conditions. People exhibiting a low degree of trust in the national government were more likely to vote in favour of the bailout, while respondents who trusted the government were more likely to vote against. Even though the question on trust made no reference to a specific national government, it is likely that respondents associated their answers with the incumbent one, which supported the “No” campaign.
Looking at individual characteristics, there were no discernible differences in voting behaviour in terms of gender or education, but age did play a role: older voters were more likely to vote “Yes” to the bailout proposal. In addition, families with a higher level of material well-being and savings were more likely to vote in favour of the measures, likely spurred by capital controls that were introduced a few days before the ballot and by the risk that Greece would have exited the eurozone had a “No” result prevailed.
Further analysis of voting patterns shows that, predictably, the respondents who voted for SYRIZA (the ruling party that invoked the referendum) in the January 2015 elections were more likely to follow the government’s direction and reject the package. Similarly, those who voted for the right-wing ANEL party, the junior partner in SYRIZA’s coalition government, were likely to vote against the proposed conditions. By contrast, supporters of New Democracy and PASOK, whose leaders had campaigned to accept the measures, were more likely to vote in favour of the bailout proposal.
SOURCE: LiTS III (2016) and author’s calculations.
NOTE: The chart shows the percentage of Greek respondents who list any of the reported actors as responsible for the economic crisis.
| (1) | (2) | (3) | (4) | (5) | (6) | |
|---|---|---|---|---|---|---|
| Respondent voted ‘Yes’ | Respondent voted ‘No’ | |||||
| Male respondent | -0.045 | -0.013 | -0.029 | 0.077 | 0.073 | 0.122 |
| (0.112) | (0.115) | (0.115) | (0.093) | (0.089) | (0.087) | |
| Respondent is 18-24 years old | -0.355 | -0.353 | -0.347 | 0.127 | 0.312* | 0.148 |
| (0.274) | (0.271) | (0.252) | (0.181) | (0.173) | (0.173) | |
| Respondent is 40-64 years old | 0.265** | 0.340*** | 0.376*** | 0.080 | 0.078 | 0.061 |
| (0.125) | (0.111) | (0.119) | (0.120) | (0.115) | (0.119) | |
| Respondent is 65 years old or above | 0.178 | 0.315* | 0.351* | -0.250* | -0.360** | -0.372*** |
| (0.207) | (0.183) | (0.180) | (0.150) | (0.145) | (0.133) | |
| Ethnic minority20 | -0.908*** | -0.959*** | -1.093*** | -1.756*** | -1.631*** | -1.790*** |
| (0.283) | (0.258) | (0.258) | (0.222) | (0.231) | (0.236) | |
| Mainland Greece | 0.587*** | 0.193 | 0.369** | -0.619*** | -0.516** | -0.616*** |
| (0.172) | (0.164) | (0.168) | (0.222) | (0.214) | (0.207) | |
| Respondent completed secondary education | -0.204 | -0.136 | -0.117 | 0.190 | 0.210 | 0.241 |
| (0.165) | (0.163) | (0.158) | (0.153) | (0.146) | (0.152) | |
| Respondent completed tertiary education | -0.018 | 0.163 | 0.153 | -0.079 | -0.104 | -0.067 |
| (0.175) | (0.184) | (0.172) | (0.168) | (0.165) | (0.163) | |
| Household had savings in 2010 | 0.243** | -0.156* | ||||
| (0.118) | (0.093) | |||||
| Household can afford a one-week holiday each year and meat, chicken or fish every second day | 0.104 | -0.032 | ||||
| (0.115) | (0.107) | |||||
| Household can meet unexpected expenditures | 0.636*** | -0.085 | ||||
| (0.114) | (0.092) | |||||
| The EU is responsible for the economic crisis | -0.304*** | 0.025 | ||||
| (0.109) | (0.113) | |||||
| Trust the national government | -0.418** | 0.644*** | ||||
| (0.175) | (0.169) | |||||
| Observations | 1329 | 1255 | 1352 | 1329 | 1255 | 1352 |
| Pseudo R2 | 0.111 | 0.051 | 0.056 | 0.095 | 0.080 | 0.100 |
SOURCE: Life in Transition Survey III (2016) and author’s calculations.
NOTE: This table reports the results of probit regressions. Standard errors in parentheses are clustered at the PSU level. *, **, and *** denote values that are statistically significant at the 10, 5 and 1 per cent levels, respectively. The reference categories for the age and education of the respondent are “Respondent is 25-39 years old” and “Respondent completed either no education or primary education”. The specifications reported in columns 1 and 4 include three controls for whether the household owns a car, a computer and a washing machine; none of them is statistically significant. The specifications reported in columns 2 and 5 include two controls for whether the respondent believes that the IMF and the ECB are responsible for the economic crisis; none of them is statistically significant. The specifications reported in columns 3 and 6 include two controls for whether the respondent trusts the national political parties and the parliament; none of them is statistically significant.
Around
70%
of Greek respondents think political parties are to blame for the economic crisis.
The chapter finds that the impact of the economic crisis on Greek households has been severe. While the transmission channels at play were similar to the ones of other countries in the period 2008-10, the consequences were more dramatic and widespread. Over 92 per cent of Greek respondents believe that the crisis has affected them “a fair amount” or “a lot”. Moreover, during the period 2010-16, 76 per cent of households suffered a negative income shock such as reduced wages or pensions, job losses, delayed or suspended wages and decreased working hours. The impact was heterogeneous, with families where the household head has either no education or only a primary education or where the household head is a woman, bigger families and ethnic minorities more likely to report having been affected by the crisis.
The crisis impacted on the lives of respondents mainly through reduction in wages or pensions, job losses and shorter working hours. The families that could respond to the economic challenges in a flexible way, either by obtaining an additional job or increasing their number of working hours, report having been less affected by the crisis. A comparison of the coping strategies adopted by Greek respondents and those of over 38,000 households in other countries during the 2008-10 period shows how Greeks have had to resort to cutting the consumption of necessities, non-necessities and services and to postponing the payment of utilities to a much greater extent. Importantly, the analysis shows that the type of coping strategy adopted by Greek families crucially depended on the gender and the education level of the household head. Female-headed households cut their consumption of necessities substantially more often than male-headed ones, while the opposite was true for non-necessities. At the same time, a lower portion of households whose head completed at least secondary education reduced their consumption of necessities in comparison to households where the head had no education or only primary education. Moreover, the availability of savings from the pre-austerity period proved fundamental in helping families cope with financial hardship more easily and avoiding cutbacks on everyday essential items such as staple foods and medicines.
The protracted crisis has not only affected the consumption levels of households, but also their general satisfaction with life and their confidence in political institutions. Today, only 1 in 10 Greeks are satisfied with their financial situation and only 24 per cent of respondents say they are satisfied with their life in general, as opposed to 72 per cent in Germany and 42 per cent in Italy. Importantly, comparing the level of material well-being of households that reported having being affected by the crisis with those that were unaffected shows that the latter have a level of material well-being at least comparable to, if not higher than, that of Germany and Italy. The same is true when the level of financial satisfaction and life satisfaction of affected and unaffected respondents are contrasted: unaffected Greek households are as satisfied as their counterparts in Germany and Italy, whereas affected ones are worse off.
While distrust of the national political institutions is widespread, the Greek public seems as satisfied with the performance of their local and regional governments as their counterparts in the transition region and in Germany and Italy. Trust levels for these institutions as well as for the ones of a national relevance are, however, lower in Greece than in the other countries. Traditional political parties and the international institutions that have negotiated the bailout terms are widely regarded as the ones responsible for the economic crisis.
In conclusion, this chapter paints a rather bleak picture of the current economic situation faced by the country. When asked where they place themselves on a 10-step income ladder today and in the future, only 16 per cent of Greek respondents believe that their position will improve in the next four years. The corresponding figures for the transition region, Germany and Italy are 48, 35 and 23 per cent respectively. This signals that, despite the recent political changes and attempts at economic reforms that have taken place in the country, Greeks do not see their situation improving for the foreseeable future.
A. Adsera, F. Dalla Pozza, S. Guriev and L. Kleine-Rueschkamp (2016)
“The impact of transition on well-being” in European Bank for Reconstruction and Development, Transition Report 2016-17, pp. 30-43.
B. Bidani, M. Fatou, and S. Zaidi (2011)
“Impact of the global economic crisis” in European Bank for Reconstruction and Development, Life in Transition: After the crisis, pp. 8-17.
A. Deaton (2008)
“Income, health, and well-being around the world: evidence from the Gallup World Poll”, Journal of Economic Perspectives, Vol. 22.2, pp. 53-72.
European Bank for Reconstruction and Development (2016)
25 years of the EBRD: The people’s view on transition. See http://bit.ly/2evwnWf (last accessed 26 October 2016).
J. Fitzgerald and J. Wolak, (2016)
“The roots of trust in local government in western Europe”, International Political Science Review, Vol. 37.1, pp. 130-46.
P. Grosjean, F. Ricka, and C. Senik (2011)
“The intangible transition: support for markets and democracy after the crisis” in European Bank for Reconstruction and Development, Transition Report 2011, pp. 60-75.
M. M. Lokshin and R. Yemtsov (2004)
“Household strategies of coping with shocks in post-crisis Russia”, Review of Development Economics, Vol. 8.1, pp. 15-32.
D. W. Sacks, B. Stevenson, and J. Wolfers (2010)
“Subjective well-being, income, economic development and growth”, National Bureau of Economic Research, NBER Working Papers 16441. See http://www.nber.org/papers/w16441 (last accessed 26 October 2016).
B. Stevenson and J. Wolfers (2008)
“Economic growth and subjective well-being: reassessing the Easterlin paradox”, Institute for the Study of Labor, IZA Discussion Papers 3654. See ftp.iza.org/dp3654.pdf (last accessed 26 October 2016).
B. Stevenson and J. Wolfers (2011)
“Trust in public institutions over the business cycle”, American Economic Review, Vol. 101.3, pp. 281-87.