Policy Research Working Paper 9581 Living Standards of Tunisian Households in the Midst of the COVID-19 Pandemic Federica Alfani Dorra Dhrif Vasco Molini Dan Pavelesku Marco Ranzani Poverty and Equity Global Practice March 2021 Policy Research Working Paper 9581 Abstract The COVID-19 pandemic has had unprecedented negative levels among the respondents, labor income among wage socioeconomic effects on the lives of millions of people workers and particularly the self-employed is still below across the world, particularly among the most vulnerable pre-pandemic levels. More than half of the households groups. The COVID-19 outbreak has exacerbated the interviewed report a worsening of their living standards issues countries were facing before the pandemic such as relative to before the start of the pandemic, and for about the unequal access to basic services, markets, labor, and cap- 40 percent of the poorest, welfare levels have continued to ital. Using five rounds of high-frequency telephone surveys deteriorate. In addition, price increases and a reduction in collected by the Tunisian National Institute of Statistics in remittances threaten to undo the progress that has been collaboration with the World Bank, this paper analyzes the achieved in raising living standards. While waiting for the deterioration in households’ welfare due to COVID-19, economy to rebound, the most vulnerable households will focusing on changes in the labor market. The results show continue to need income support. that although employment has now rebounded to pre-crisis This paper is a product of the Poverty and Equity Global Practice. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at vmolini@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Living Standards of Tunisian Households in the Midst of the COVID-19 Pandemic Federica Alfani • Dorra Dhrif ∗ Vasco Molini •1 Dan Pavelesku• Marco Ranzani • JEL: I31, J21, J33. Keywords: Welfare, Recessions, Labor market, Coronavirus, Tunisia. • MENA Poverty and Equity Global Practice. ∗ Institut National de la Statistique of Tunisia. 1 Corresponding author at vmolini@worldbank.org; Federica Alfani f.alfani@worldbank.org; Dorra Dhrif dhrif.dorra@ins.tn; Marco Ranzani mranzani@worldbank.org; Dan Pavelesku dpavelesku@worldbank.org 1. Introduction The COVID-19 outbreak has had unprecedented negative socioeconomic effects on the lives of millions of people across the world, particularly among the most vulnerable groups. 1 Tunisia has had to endure the human and economic costs of the COVID-19 pandemic, and the difficulties brought on by containment measures and restrictions to mobility. The first cases of COVID-19 in Tunisia were reported on March 2, 2020; as of January 13, 2021, a total of about 165,000 cases and over 5,300 deaths have been recorded by Tunisian authorities. Following the pattern seen in other countries, the virus spread slowed between May and July, but cases rose again in the autumn, with peaks in October and November. The public health emergency during the height of the COVID-19 pandemic and the introduction of lockdown measures led to a 10 percent contraction of the Tunisian economy in the first nine months of 2020 (Institut National de la Statistique, INS); for its part, the World Bank (2020a) projects that the country’s economy will contract by 9.2 percent in 2020.2 The pandemic, the containment measures and restrictions to mobility have led to: (i) steep job losses and drops in incomes; (ii) price increases; (iii) decline in other income sources; and (iv) disruption in the delivery of health and education services. Combined, the health and socioeconomic crisis of 2020 could reverse some of the progress Tunisia has made in poverty reduction and raising living standards. Monitoring the socioeconomic impact of the crisis has been challenging in the context of the COVID- 19 pandemic; several national statistical offices have resorted to telephone surveys to overcome the challenge. Findings from the recently released COVID-19 High Frequency Survey Global Dashboard 3 1 Based on data collected by Johns Hopkins University, as of January 13, 2020, over 91 million cases have been confirmed worldwide, along with almost 2 million deaths. 2 Estimates of Tunisia GDP growth rates during the first nine months of 2020 are available at: http://www.ins.tn/fr/themes/compte- de-la-nation#2914. 3 COVID-19 High Frequency Survey Monitoring Dashboard provides 93 harmonized indicators on 14 topics, allowing users to compare and analyze how COVID-19 impacts vary across countries over time and by industry sector and regions. Data can be downloaded for further analysis at https://www.worldbank.org/en/data/interactive/2020/11/11/covid-19-high-frequency- monitoring-dashboard 2 show that widespread impacts amplify pre-existing inequalities between rich and poor countries and between the haves and the have-nots within countries (Sanchez-Paramo and Narayan, 2020). In Tunisia, between March and October 2020, the INS, in collaboration with the World Bank, launched five rounds of high-frequency telephone household surveys to assess the socioeconomic impact of the COVID-19 pandemic on a nationally representative panel of about 1,000 households. These results have been initially published in separate notes available on the INS website. 4 The rapid expansion of telephone surveys on the impact of COVID-19 on households has been accompanied by a flurry of studies and research papers which have simulated the economic impacts of the COVID-19 shock on poverty and welfare. Papers on this include Sumner et al. (2020) and Morsy et al. (2020), ElKadhi et al. (2020), ITES, (2020); and Kokas et al. (2020) in the case of Tunisia. 5 This literature typically identifies employment loss as the major channel of impact on household welfare and poverty (Josephson et al., 2020; Headey et al., 2020). Cross-country research on the effects on the labor market shows that the magnitude of the impact of the COVID-19 shock differs, depending on the institutional context, economic structure and work schemes in place; it particularly affects tasks that cannot be carried out remotely, or by less educated workers, the youth, women, and the self-employed (Adams-Prassl et al., 2020; Alon et al., 2020; Bartik et al., 2020; Blundell and Machin, 2020; Cajner et al., 2020; Dingel and Neiman, 2020; Mongey and Weinberg, 2020; Montenovo et al., 2020; von Gaudecker et al., 2020). Other contributions (e.g. Alon et al., 2020) gauge the impact of COVID-19 on social norms and role models, especially in households with children, where a reallocation of duties within the household is very likely, with persistent effects on gender roles and the division of labor. In Tunisia, beyond the labor market, the COVID-19 pandemic has exacerbated the unequal access to basic services, markets, and capital, as well as contributed to increase social and regional disparities widening the gap between the coast and the interior. In particular, the contagion risk as well as the lack of access to health care in case of infection are higher for the poor and most vulnerable ones, 4 http://www.ins.tn/fr/publication/suivi-de-l%E2%80%99impact-socio-%C3%A9conomique-du-covid-19-sur-les- m%C3%A9nages-tunisiens-octobre-2020 5 No less than 100 countries have indicated that they are publishing telephone data on the World Bank website. 3 mainly due to overcrowded living conditions, low access to water at home, and higher prevalence of intergenerational households (Bentamansourt, 2020; ElKadhi et al., 2020; Kokas et al., 2020; Mahjoub, 2020). Despite the clear interest on the topic, little evidence or data are available on the socioeconomic impacts of the COVID-19 pandemic in the Middle East and Northern Africa (MENA) countries. This paper presents an analysis of five rounds of telephone surveys in Tunisia, and sheds light on the impact of COVID-19 on households’ welfare during the pandemic, and particularly on the changes that have occurred in the labor market (i.e. employment losses and reduced labor income). 6 In particular, it seeks to show how the declining living standards reported by households are continuing, despite the re-opening of the Tunisian economy following the lockdown. The survey results indicate that the combined public health and economic crises have had a sizeable negative effect on the living standards of Tunisian households, particularly among the most vulnerable (i.e. in the bottom 40 percent of the consumption distribution). About one in two households reported a decline in welfare, as compared to prior the pandemic, and about one in five households declared a further worsening throughout the pandemic. Although employment has rebounded to pre-crisis levels among survey respondents, findings show that labor income among wage workers, particularly among the self-employed, is still below pre-pandemic levels. Contributing to the deterioration in living standards are an increase in food prices and a fall in remittances. The rest of the paper is articulated as follows. In section 2, we provide detailed information on the 5 rounds of household telephone surveys, the data collection process and how it overlaps with lockdown decisions. In section 3, we briefly discuss the key transmission channels of an aggregate shock such as the COVID-19 pandemic on living standards and then present results of the analysis of changes in living standards among Tunisian households throughout the pandemic with a focus 6 Summary notes on the Enquête téléphonique auprès des ménages pour étudier et suivre l’impact du COVID-19 sur le quotidien des Tunisiens, monitoring the socio-economic impact COVID-19 on Tunisian households are available on the INS website at the following link: http://www.ins.tn/fr/recherche-publication. 4 on the labor market as one of the driving forces behind the deterioration in households’ welfare. We provide concluding remarks and policy recommendations in section 4. 2. Data The first COVID-19 cases appeared in Tunisia at the beginning of March 2020. On March 20, the government announced a strict lockdown, and on May 4, it launched the first of three phases of stepwise deconfinement measures, resulting in the gradual reopening of the economy between May 4 and June 14, 2020. The Institut National de la Statistique (INS), in collaboration with the World Bank, launched a series of five surveys to study and monitor the socioeconomic impact of COVID-19 on the daily lives of Tunisians. The first round was conducted from April 29 to May 8, 2020 – at the end of the strict lockdown and the beginning of the first reopening stage (Figure 1). The second round took place between May 15 and 21, the third round about 2 weeks later, between June 8 and 15. The fourth round occurred between June 22 and 30, and the fifth took place between October 4 and 16, about 3-4 months after the end of the lockdown (Figure 1). Figure 1. Timeline of Survey Data Collection and Lockdown Measures 5 The surveys were administered by telephone to a panel of 1,339 households, a subsample of the 2015 Enquête nationale sur le budget, la consommation et le niveau de vie des ménages (EBCNV-2015).7 The sample makes it possible to obtain statistically significant results at the national level, as well as by rural/urban areas and households’ consumption quintiles. 8 In the first round, a total of 1,030 households responded to the telephone survey, giving a response rate of 77 percent. In the second round, the response rate fell to 67 percent, 63 percent in the third, 59 percent in the fourth and 53 percent in the fifth (Table A 2 in the Annex). The questionnaire for all five rounds included sections on health care preventive measures and social behavior and economic activity (Table A 1 in the Annex). Specific modules aimed at capturing particular sets of information were developed: rounds 1 and 2 sought responses on access to goods and services and preventive measures and aid; rounds 2-5 contained modules on the financial situation of households; rounds 2 and 5 looked into the impact of COVID-19 on children’s education; rounds 3 and 4 queried attitudes on targeted confinements; and rounds 4 and 5 examined access to health services. 7 The sample comprises households interviewed in the 2015 Enquête nationale sur le Budget, la Consommation et le Niveau de vie des ménages that reported their phone numbers. 8 The construction of sampling weights entailed: (i) applying the inverse of the selection probability to adjust for the subsample selection process; and (ii) to adjust for non-response, modifying the design weights by a factor inversely proportional to the response rate within 12 homogeneous groups selected based on the propensity score logit model. Twelve groups with an average of about 111 units per group were formed by: (i) two groups based on deciles of consumption expenditures (top 40 percent and bottom 60 percent); (ii) two groups based on urban/rural location of residence; and (iii) three groups based on the region where they reside, i.e. northern, central or southern regions of the country. 6 3. The Impact of the COVID-19 Pandemic on the Living Standards of Households This section describes how the living standards of Tunisian households have changed during the pandemic and investigates labor market changes as one of the potentially driving forces behind the deterioration in households’ welfare. The COVID-19 outbreak has had unprecedented negative socioeconomic effects on the lives of millions of people across the world, particularly among the most disadvantaged and vulnerable. The key mechanisms through which aggregate shocks, such as the COVID-19 pandemic and the consequent economic recession, are transmitted and impact households’ welfare are through four main channels, namely: (i) income from labor; (ii) income from other sources; (iii) prices; and (iv) service disruption (Figure 2). Figure 2. Key transmission channels of short-term impact of the COVID-19 shock on households’ welfare • The labor market is a key transmission channel and is particularly important in countries where the poor and vulnerable rely on their labor to make ends meet. Private sector firms experienced 7 a triple shock following the COVID-19 outbreak. On the supply side, the lockdown and restrictions to mobility closed businesses completely, and created difficulties in accessing labor and other inputs. On the demand side, declining incomes reduced consumption to essential basic needs. Third, some Tunisian enterprises were not supported by their banks, nor were they able to benefit from government assistance during the pandemic, leading to financial distress for many of them. Limited mobility and the unavailability of home-based work, especially for low- skilled workers performing tasks that cannot be done remotely, has translated into job losses, reduced earnings, and lower living standards. • Other sources of income, particularly public and private transfers, can potentially be impacted by the pandemic. Decreasing levels of international (and domestic) remittances are the byproduct of economic contraction in those countries/regions where emigrants are working. In countries with limited fiscal space, any reduction in the volume of public transfers, and/or in the size of the target populations following an economic downturn, can contribute to deteriorating living standards. On the other hand, public transfers could be increased following an economic downturn, and could therefore help mitigate the impact of a crisis. • Another effect of the pandemic on welfare can occur via changes in prices and/or shortages of basic food items and medicines due to disruptions in international and national supply chains, or domestic production contractions. On a positive note, falling commodity prices, e.g. oil and gas, can offset some of the price increases on other goods. • Finally, the paucity and low quality of existing Internet connections have a negative impact on households’ welfare as it limits access to education and distance learning. The saturation of health care systems during the height of hospitalizations of COVID-19 patients can make it difficult to access health or medical services, particularly in cases where such services are weak to begin with. Self-reported changes in living standards When looking at the self-reported changes in living standards (Figure 3), one in two households reported that COVID-19 outbreak has led to a deterioration in living standards at the time of the 8 interview, as compared to the period prior to the COVID-19 outbreak, i.e. March 2020. In May, 54 percent of the interviewees reported a decline in their welfare; this share declined to 49 percent in the first half of June, and to 43 percent in the second half of the month; in the first half of October, it rose again to 46 percent. Figure 3. Household-level self-reported change in living standards, compared to before the start of the pandemic, by survey round Round 5 14.3 31.6 50.0 3.7 0.3 Round 4 10.9 32.0 50.4 6.6 0.2 Round 3 15.4 33.9 46.6 4.1 Round 2 21.7 32.4 42.3 3.5 0 20% 40% 60% 80% 100% percent Much worse Worse Same Better Much better Source: Based on data from the Enquête téléphonique auprès des ménages pour étudier et suivre l’impact du COVID19 sur le quotidien des Tunisiens, rounds 1-5, INS and World Bank. Living standards deteriorated particularly among the poor and the bottom 40 percent compared to the rest of the population (Figure 4). 9 In May, about 67 percent of households in the lowest quintile reported that their living standards had worsened, compared to the situation they faced in March; in the second lowest quintile the share is estimated at 68 percent among households, and declined to 57, 47 and 37 percent, respectively, among household in the middle, the fourth and fifth quintile of the household consumption distribution (Figure 4). Similarly, in the following rounds, the 9 The quintiles are based on the distribution of consumption expenditures in the 2015 Household Budget Survey. In 2015, the headcount poverty rate was estimated at 15.2 percent. 9 percentage of households declaring that their living standards were worsening is consistently higher at the bottom of the consumption distribution than at the top (Figure 4). Figure 4. Household-level self-reported change in living standards, compared to before the start of the pandemic, by pre-pandemic quintile and survey round Source: Based on data from the Enquête téléphonique auprès des ménages pour étudier et suivre l’impact du COVID-19 sur le quotidien des Tunisiens, INS and World Bank. In the final three survey rounds, households were also asked about their welfare, as compared to the previous month, i.e. once the pandemic had already had an impact on living standards. Twenty- four percent of households reported a decline in their living standards in the first half of June, as opposed to the month before (Figure 5). This share decreased to about 15 percent in the second half of June as the lockdown was lifted, but rose again to 25 percent in the first half of October. This evidence points to a severe decline in household welfare as the pandemic unfolded and extended well beyond the end of restrictions to individual mobility due to a deepening economic downturn. 10 Similar to the previous response, the decline is larger at the bottom of the consumption distribution. For example, compared to the month before the survey, in the October survey round about 38 percent of households in the first quintile reported a decline in welfare; this compares to 24 percent of households in the second quintile, 29 percent in the middle of the distribution, and 22 and 16 percent in the fourth and fifth quintile, respectively (see Figure A1 in the Annex). Figure 5. Household-level self-reported change in living standards relative to the month before the interview by survey round Round 5 6.7 18.7 62.9 11.0 0.7 Round 4 4.2 10.6 56.1 28.7 0.3 Round 3 7.1 16.5 55.4 20.9 0.1 0 20% 40% 60% 80% 100% percent Much worse Worse Same Better Much better Source: Based on data from the Enquête téléphonique auprès des ménages pour étudier et suivre l’impact du COVID19 sur le quotidien des Tunisiens, INS and World Bank. Changes in the labor market The labor market seems the most obvious transmission channel of aggregate shock to household welfare. This is particularly true for the poor and the bottom 40 percent in Tunisia as they tend to live off wages and business income in the private sector. Indeed, the telephone surveys showed that the deterioration in living standards appeared to be due to: (i) a decline in employment, particularly during the lockdown; and (ii) a reduction in income from labor, which seems to continue as individuals go back to work, particularly among the self-employed. 11 Figure 6. Share of respondents who report being employed, by survey round 100.0 80.0 59.0 58.5 60.0 52.2 Percent 38.0 40.0 23.2 20.0 0.0 Round 1 Round 2 Round 3 Round 4 Round 5 Source: Based on data from the Enquête téléphonique auprès des ménages pour étudier et suivre l’impact du COVID19 sur le quotidien des Tunisiens, INS and World Bank. In May, in the midst of the lockdown, employment among respondents dropped to 23 percent - (Figure 6). This compares with an estimated employment ratio in 2017 of about 62 percent among household heads (using labor force survey data). Figure 7. Share of private sector employees who report being paid their full salary, by survey round 100.0 88.5 83 82.6 80.0 62.8 60.0 Percent 40.0 20.0 1.6 0.0 Round 1 Round 2 Round 3 Round 4 Round 5 Source: Based on data from the Enquête téléphonique auprès des ménages pour étudier et suivre l’impact du COVID19 sur le quotidien des Tunisiens, INS and World Bank. 12 With the lockdown coming to an end and the re-opening of most economic activities, the share of respondents who had worked in the week preceding the interview rose from 38 percent in the latter half of May to 52 and 58 percent in June and October, respectively. Despite this clear rebound to pre-crisis levels among respondents, not everybody went back to work. Data from the Tunisia labor force survey indicate that total employment in the third quarter of 2020 was still 1.5 percent below the level estimated in the first quarter. In other words, about 54,000 jobs had been shed between quarters 1 and 3. Figure 8. Distribution of self-employed by degree of change in business income, as compared to the 2 weeks before the interview and distribution of main reason for the decline, by survey round a. Distribution of self-employed by change in business income relative to the 2 weeks before interview 13 b. Distribution of main reason for the decline in business income among the self-employed Source: Based on data from the Enquête téléphonique auprès des ménages pour étudier et suivre l’impact du COVID19 sur le quotidien des Tunisiens, INS and World Bank. Unlike employment, incomes have not bounced back to the level observed before the onset of the COVID-19 pandemic. At the start of the re-opening, a rising share of private sector employees reported being paid a full salary. This share rose from 63 percent in the latter half of May to 83 percent in June with the full re-opening (Figure 7). As of October 2020, about 89 percent of private sector employees were reporting having been paid a full wage. Labor income appears to have deteriorated considerably more among the self-employed. The share of self-employed reporting a reduction in business income, as compared to the two weeks before the survey, declined from 63 percent in early May to 56 percent in the latter half of May, and from 43 and 24 percent in the early and latter half of June, respectively (Figure 8-panel a). By October 2020, this share has increased to about 28 percent. The main factor cited by respondents for their fall in income was a lack of customers since the re-opening (from 28.4 percent in late May to 47.6 percent in October). The lockdown and subsequent closure of workplaces in May was the main reason for the reduction in income of about one in two of the self-employed (Figure 8-panel b). A linear probability model at the household level is estimated to simultaneously control for the effects of the factors described above, as well as for additional individual characteristics on changes in households’ living standards. The dependent variable is a dummy that equals one if a household 14 has reported a deterioration in living standards relative to the month before the interview. Controls are included for the sex, age group and educational level of household heads (ℎ ), for a respondent’s labor force status, and conditional on being employed for her employment type, and for changes in her labor income compared to the period before the pandemic (ℎ ) (equation 1). We also control for the pre-pandemic distribution of household consumption expenditures (ℎ ) and we add fixed effects for each survey round ( ). ℎ = + ′ ℎ + ′ ℎ + ′ℎ + + ℎ (1) Table 1 reports the estimates for two specifications. The first specification (M1) does not include any details on household heads’ characteristics, and only contains variables measured at the time of the survey on respondents’ labor market status and change in income. The second specification (M2) adds gender, age and educational level of household heads. 10 The estimates do not differ significantly between the two specifications. Table 1. Correlates of the probability of posting a deterioration in living standards, as compared to the month before the interview – the linear probability model M1 M2 Male household head 0.066** Age group 35-44 -0.100** 45-64 -0.113*** 65+ -0.132*** Educational level Primary education -0.063** Secondary education -0.071** Tertiary education -0.109*** Employment type Not employed 0.439*** 0.429*** Private sector employee -0.049* -0.072** 10 A household head refers to an individual who was the head of the household at the time of the 2015 household budget survey. Any changes within households, including deaths, marriages, divorces, migration, etc., might have altered households’ structure between 2015 and 2020. For this reason, we also estimate a regression that does not include household heads’ demographics. We do control for the quintile of the pre-pandemic household welfare that partially captures returns to household heads’ characteristics. By the same token though, the position of each household along the distribution of consumption expenditures might have changed since 2015. However, this is the best information available. 15 Self-employed/employer/contributing family worker 0.371*** 0.329*** Change in wages Partial salary 0.388*** 0.385*** No salary 0.195* 0.195* Change in business income Business income lower than usual/no business 0.292*** 0.294*** Quintile of household consumption 2nd quintile -0.135*** -0.124*** 3 quintile rd -0.112*** -0.095*** 4th quintile -0.189*** -0.166*** 5 quintile th -0.253*** -0.215*** Survey round Round 4 -0.063*** -0.062*** Round 5 0.043** 0.045** Constant -0.031 0.077 Observations 2,323 2,323 R-squared 0.095 0.104 Reference category: 15-34; no education; public sector employee; full salary, business income as usual or more than usual; Quintile 1; Round 3. Robust standard errors in parentheses. Statistical significance: *** p<0.01, ** p<0.05, * p<0.1. Source: Based on data from the Enquête téléphonique auprès des ménages pour étudier et suivre l’impact du COVID19 sur le quotidien des Tunisiens, INS and World Bank. The probability of a respondent declaring a worsening in living standards is positively correlated with household head having a low educational level and being in the youngest age group (15-34). For example, a household headed by someone aged 35-44 is 10 percent less likely to see a deterioration in living standards the month before the interview as compared to a household headed by someone aged under 34. Similarly, households headed by someone with a primary (-6.3 percent), secondary (-7.1 percent) and tertiary (-10.9 percent) education are less likely to report a deterioration in living standards, as compared to a household headed by someone with no education. Individuals not employed at the time of the survey, the self-employed and contributing family workers have a higher probability of reporting a deterioration of their living standards (42.9 and 32.9 percent, respectively), as compared to someone working as a civil servant. 16 Private sector wage workers have a lower likelihood (-7.2 percent) than civil servants of experiencing worsening living standards. As reported by respondents, the magnitude of the coefficients attached to the changes in labor income, both in wages and business income is sizeable. A private sector employee receiving a partial salary, or no salary at all, has a higher probability of reporting lower living standards than a civil servant receiving a full salary (38.5 and 19.5 percent, respectively). Similarly, respondents employed as non-wage workers with lower than usual, or no business income at all, have a higher likelihood (29.4 percent) of experiencing lower living standards, as compared to a civil servant receiving a full salary. Finally, the household welfare measured before the pandemic, as captured by the quintile on household consumption expenditures in 2015, is highly statistically significant. More affluent households, compared with households in the first quintile, have a lower likelihood of reporting a deterioration in living standards during the pandemic after controlling for age, gender and educational level of household heads, as well as for their labor market status and reported changes in labor income. These households tend to have more access to savings or credit which, in turn, enables them to withstand economic shocks better than poorer households. Changes in prices Other mechanisms can contribute to the transmission of an aggregate shock to households’ welfare; one of these mechanisms – price changes – can directly impact consumption levels and may force some households to adopt negative coping strategies by reducing their consumption of specific items. The consumer price index (CPI) series constructed by the Tunisia National Statistics Institute shows that, in October 2020, the total CPI rose by 4 percent, as compared to January 2020; over the same period in 2019, the CPI increased by 4.5 percent (Figure 9). Nonetheless, the price increase observed during the lockdown months of March-May is slightly higher in 2020 than in 2019. In addition, the price index of food items increased considerably more between January and October 2020 (+4.1 percent), compared to the same period in 2019 (+2.9 percent). Third, the CPI trend for food items shows a first acceleration during the lockdown, 17 followed by a second one beginning in September. As less affluent households typically spend a larger share of their consumption on food items (about 40 percent among households in the bottom 40 percent, compared with 35 percent among the rest of the population), this price dynamic penalizes more households at the bottom of the consumption distribution. Figure 9. Consumer Price Index Trends, relative to the first month of January 2019 and 2020 5.0 Total Food and non-alcoholic beverages 4.5 4.0 3.5 3.0 Percent 2.5 2.0 1.5 1.0 0.5 0.0 Jul-19 Jul-20 Jun-19 Jun-20 Apr-19 May-19 Apr-20 May-20 Aug-19 Aug-20 Jan-19 Feb-19 Mar-19 Sep-19 Jan-20 Feb-20 Mar-20 Sep-20 Oct-19 Oct-20 Source: Based on data from the Consumer Price Index, INS. Changes in remittances Other sources of income, particularly remittances, can impact the living standards of households during a downturn. About 1.22 million Tunisians lived abroad in 2012, largely in Europe; these expatriates contribute to sustaining the living standards of family members in Tunisia through remittances. Although no information on the value of remittances is available at the household level, the World Bank (2020b) estimated that remittances valued at $ 1.9 billion accounted for 4.9 percent of the country’s GDP in 2019. Due to the COVID-19 pandemic, the World Bank projects a sharp drop in remittances of about 15 percent in 2020, which contributed to a deterioration of living standards, particularly among households that had benefitted the most from this source of income (World Bank, 2020c). 18 Public transfers, and particularly social protection systems, can provide a safety net for those who lose their job to support their living standards, and give them the means to look for new employment. Although Tunisia has a comprehensive social insurance system in place, high unemployment and levels of informality has created low coverage rates; in addition, unconditional cash transfers are poorly targeted (OECD, 2015). Shortly after the COVID-19 outbreak and subsequent lockdown, the Government of Tunisia introduced short-term work schemes which included : (i) a wage subsidy of DT 200 per month in April and May, which enterprises could access by applying through a dedicated platform (helpenterprise.social.tn); and (ii) a one-off cash transfer of DT 200 for micro-enterprises under a special tax regime (forfaitaire).11, 12 About 110,000 micro-enterprises received the transfer, and most firms under this regime are self- employed or have a maximum of 1-2 employees. In addition, 623,000 vulnerable households received benefits from the Programme national d'aide aux familles nécessiteuses (PNAFN), and discounted health care cards (Aide medical à tarif réduit, AMG2 program) have been supported through two cash transfers of equal amount of DT 50 and DT 200, respectively. 13 Furthermore, a one-off cash transfer of DT 200 was credited to families hosting an elderly person, fostering children without parental support, and those hosting a person with handicap. Persons with low retirement pensions (less than DT 180) received an exceptional monetary payment of DT 100 in April 2020. Moreover, a one-off cash transfer (DT 200) was introduced for an additional 300,000 poor and vulnerable households working in the informal sector. 14 11 The government has recently announced the re-opening of the platform to provide support to employees in the tourism sector. 12 Micro-enterprises could apply through a dedicated platform (batinda.gov.tn). 13 In the case of the AMG1 program, the first transfer corresponded to DT 50 in April 2020, while the second amounted to DT 60 in May 2020. 14 Information on compensatory measures targeted to Tunisian households has been drawn from “La Tunisie face à la covid-19 à l’horizon 2025. Scénarios fin 2020 et stratégie de sauvetage et de sortie de crise : quels équilibres sociaux?”, Institut Tunisien des Études Stratégiques. Available at: https://www.realites.com.tn/wp-content/uploads/2020/11/Letude-Sociale-de-lITES-est-accessible-via- ce-lien..pdf 19 While these temporary compensation schemes and one-off transfers have ended, the economy is still far from a rebound. The economic contraction has meant that many Tunisian households, particularly private sector employees and the self-employed working in the sectors that have been most affected by the crisis, e.g. construction, manufacturing, accommodation and food services activities and transport, will continue to experience deteriorating living standards. Strengthening and adequate targeting of social protection programs could help vulnerable households make ends meet until the economy recovers. The latest World Bank projections indicate that the Tunisia economy will contract by 9.2 percent in 2020, with a large reduction in the services sector (World Bank, 2020a), and it is not expected that the economy will bounce back to a pre-crisis level before 2022. 4. Conclusion The COVID-19 outbreak and the economic crisis that followed have had a negative effect on the living standards of many people around the world. Five consecutive rounds of telephone interviews, conducted by the INS during and immediately after the lockdown, presented an opportunity to evaluate the impact of Covid-19 on Tunisian households and identify some of the transmission channels of the global pandemic on the daily life of Tunisian people. Survey results indicate that living standards deteriorated in about one in two households, as compared to the period before the pandemic, and for over six in ten among households in the bottom 40 percent. Moreover, about one in five households reported experiencing worse living standards during the pandemic. Among the possible transmission channels, the labor market played an important role. Although COVID-19 has had negative effects on everyone’s welfare, poor and vulnerable households are particularly hit because of their unequal access to basic services, especially heath care in case of infection, as well as coping mechanisms to put in place. While survey respondents reported that employment had rebounded to pre-crisis levels, wage workers and the self-employed indicate that their labor income is still below the levels observed 20 prior to the pandemic. Food price increases and a sharp reduction in remittances have also contributed to the deterioration of living standards. The government’s immediate response to the crisis has included compensation schemes for private sector employees and an income support scheme for micro-enterprises; both schemes have played an important role in limiting job destruction and income loss. Income support will continue to be needed for the most vulnerable households following the end of the emergency programs, but also because an economic recovery is not expected for another one or two years. Particular attention will need to be paid to households that rely more than others on remittances, as well as those households with private sector employees and self-employed working in the sectors most affected by the crisis. These households, along with the poorest, are also bearing the brunt of increasing food prices, and will continue to experience deteriorating living standards due to a lack of aggregate internal and external demand. In the coming months, particularly affected households will need to be supported through targeted social protection schemes. 21 References Adams-Prassl, A., T. Boneva, M. Golin, C. Rauh. 2020. Inequality in the impact of the coronavirus shock: Evidence from real time surveys, Journal of Public Economics, Volume 189. Alon, T., M., Doepke, J., Olmstead-Rumsey, and M., Tertilt. 2020. The impact of COVID-19 on gender equality, NBER Working Paper 26947. Bartik, A.W., M., Bertrand, Z.B., Cullen, E.L., Glaeser, M., Luca, and C.T., Stanton. 2020. How are small businesses adjusting to COVID-19? Early evidence from a survey, NBER Working Paper 26989. Bentamansourt, N. 2020. Tunisie-Coronavirus : Les oubliés confinés ont soif…Une bombe à retardement? African Manager, April 2020, https://africanmanager.com/tunisie-coronavirus-les- oublies-confines-ont-soifune-bombe-a-retardement/ Blundell, Jack and Machin, Stephen (2020) Self-employment in the Covid-19 crisis: a CEP Covid-19 analysis. CEP Covid-19 Analysis (003). London School of Economics and Political Science, London. Cajner, T., L.D., Crane, R.A., Decker, J., Grigsby, John, A., Hamins-Puertolas, E., Hurst, C., Kurz, and A., Yildirmaz. 2020. The US Labor Market During the Beginning of the Pandemic Recession, NBER Working Paper 27159. Dingel, J., and B., Neiman. 2020. How many jobs can be done at home? National Bureau of Economic Research 26948. ElKadhi, Z., D. Elsabbagh, A. Frija, T. Lakoud, M. Wiebelt, and C. Breisinger. 2020. The impact of COVID-19 on Tunisia’s economy, agri-food system, and households. MENA Policy Note 5. Washington, DC: International Food Policy Research Institute (IFPRI). Institut Tunisien des Études Stratégiques - ITES. 2020. La Tunisie face à la covid-19 à l’horizon 2025. Scénarios fin 2020 et stratégie de sauvetage et de sortie de crise : quels équilibres sociaux ? Available at : https://www.realites.com.tn/wp-content/uploads/2020/11/Letude-Sociale-de-lITES- est-accessible-via-ce-lien..pdf Kokas D., G., Lopez-Acevedo, A.R., El Lahga, V. Mendiratta. 2020. Impacts of COVID-19 on Household Welfare in Tunisia. Policy Research Working Paper 9503. World Bank, Washington D.C. Mahjoub A. 2020. Pandémie COVID 19 en Tunisie : les Inegalites, les Vulnerabilites a la Pauvrete et au Chômage, Forum tunisien pour les Droits Economiques et Sociaux. Mongey, S., and A., Weinberg. 2020. Characteristics of workers in low work-from-home and high personal-proximity occupations, BFI White Paper, Becker Friedman Institute, Chicago. Montenovo, L., X. Jiang, F.L. Rojas, I.A., Schmutte, K.I., Simon B.A., Weinberg, C., Wing. 2020 Determinants of Disparities in Covid-19 Job Losses, NBER Working Paper 27132. Morsy H., L Balma, and A.N. Mukasa. 2020. ‘Not a Good Time’: Economic Impact of COVID-19 in Africa, Working Paper Series N° 338, African Development Bank, Abidjan. 22 OECD. 2015. A Reform Agenda to Support Competitiveness and Inclusive Growth. Better Policies Series, March. OECD Publishing, Paris. Sanchez-Paramo, C. and Narayan, A. 2020. Impact of covid-19 on households: what do phone surveys tell us? Voices, World Bank Blogs, 20 November 2020, [accessed 10 December 2020], https://blogs.worldbank.org/voices/impact-covid-19-households-what-do-phone-surveys-tell-us Sumner, A., C. Hoy and E. Ortiz-Juarez. 2020. Estimates of the impact of COVID-19 on global poverty, UNU-WIDER Working Paper 43/2020. Available at : https://www.wider.unu.edu/publication/estimates-impact-covid-19-global-poverty. von Gaudecker, H.-M., R., Holler, L., Janys, B., Siflinger, and C., Zimpelmann. 2020. Labour Supply in the Early Stages of the Covid-19 Pandemic: Empirical Evidence on Hours, Home Office, and Expectations, IZA Discussion Paper No. 13158. World Bank 2020a. Rebuilding the Potential of Tunisian Firms. Tunisia Economic Monitor, Fall 2020. World Bank, Washington D.C. World Bank. 2020b. COVID-19 Crisis through a Migration Lens. Migration and Development Brief 32, April. World Bank, Washington D.C. World Bank. 2020c. Phase II: COVID-19 Crisis Through a Migration Lens. Migration and Development Brief 33. October. World Bank, Washington D.C. 23 Annex Table A 1. Modules of the survey questionnaires by survey round Section Round 1 Round 2 Round 3 Round 4 Round 5 Preventive measures and social behavior ꭙ ꭙ ꭙ ꭙ ꭙ Access to goods and services ꭙ ꭙ Economic activity ꭙ ꭙ ꭙ ꭙ ꭙ Financial situation of the household ꭙ ꭙ ꭙ ꭙ Impact of COVID on children’s education ꭙ ꭙ Access to health services ꭙ ꭙ Attitudes to the targeted lockdown ꭙ ꭙ Aid ꭙ ꭙ Table A 2. Sample size and response rate by survey round Dataset Observations, Response rate, number percentage Round 1 1,032 77 Round 899 67 2 Round 837 63 3 Round 789 59 4 Round 714 53 5 24 Figure A1. Household-level self-reported change in living standards, as compared to the month before the interview, by pre-pandemic quintile and survey round Source: Based on data from the Enquête téléphonique auprès des ménages pour étudier et suivre l’impact du COVID19 sur le quotidien des Tunisiens, INS and World Bank. 25