Benefits and transfers during COVID-19

Declines in household income would have been worse had social protection measures not been implemented. From the onset of the COVID-19 crisis, the Uruguayan government implemented social protection adjustments and new measures, all of them conceived as temporary measures. As discussed in Bastagli and Lowe (Reference Bastagli and Lowe2021), the configuration of the precrisis social protection system was crucial in ensuring rapid support for existing program participants. On the benefit side, the main measures consisted of an increase in the amount of existing cash-transfer measures (Asignaciones Familiares, or AFAM, and Tarjeta Uruguay Social, or TUS), as well as intensive use of the unemployment insurance program and the implementation of a new transfer targeted at informal workers. On the tax side, a temporary income tax (COVID-19 Emergency Tax) for public workers (excluding health workers) and pensioners receiving an income above a certain threshold was introduced. This tax was levied for two months in 2020 and two months in 2021.

The TUS is an unconditional monthly cash transfer, granted through a prepaid card to households with extreme economic vulnerability. The card can be used in a network of affiliated stores to purchase food and cleaning and hygiene products (purchases with the card are exempt from the value-added tax). The monthly transfer amounts range from around US$30 for a household with no children or one child and up to US$81 for a household with four or more children; the supplement per pregnant woman or per child younger than four years of age amounts to US$7. Since 2011, households in the worst economic situation have received double the standard transfer.

The AFAM is a cash transfer focused on poor households with pregnant women or children younger than eighteen years old, conditional on certain educational and health requirements. Around 40 percent of Uruguayan children younger than eighteen are beneficiaries of the program. The amount of the transfer depends on the number of children in the household, based on an equivalence scale, and to generate positive incentives for education, higher transfer amounts are granted according to the beneficiaries’ progress in the educational system. The basic amount of the transfer is around US$53, and the supplement for secondary education attendance is around US$23. Most households receiving TUS are also beneficiaries of AFAM (94 percent). The exception is the few poorer households in the lowest 10 percent of the distribution with no child in the household.

The other social protection program that played a relevant role during COVID-19 was unemployment insurance. This program had different schemes in operation in Uruguay before COVID-19: the typical design to face a layoff event but also modalities of suspension and reduction. These modalities act like a work-sharing or shared-work program, constituting a subsidy for temporary reductions in the number of hours worked in firms that face adverse demand shocks, the objective being to avoid laying workers off.

Contrary to what happened in other countries in the region, the number of beneficiaries of the two main cash transfers, AFAM and TUS, did not increase during the COVID-19 crisis: The Uruguayan government decided to scale up benefits in these programs without increasing the programs’ coverage. Starting in April 2020, the government increased the amount of the cash transfers received by vulnerable households. The original rule was that if a household received both TUS and AFAM, the stipulated relative increase was applied to the TUS transfer amount, whereas if a household received only AFAM, the relative increase was applied to the AFAM transfer. The increases were set at 50 percent of the corresponding transfer amount from April to December 2020, except for October. Given the amounts involved in each transfer, this original design implied relevant horizontal inequalities, as poorer households received a duplicate TUS (which was less than if AFAM was duplicated). This was corrected during 2021, when all households that were receiving both AFAM and TUS received an increase of 50 percent of the amount of the AFAM transfer from January to May; from June to August, they received double the (standard) transfer amount. Finally, increases of 70 percent and 50 percent in the amount of the transfer were given in September and October, respectively. In addition, the government established that prenatal AFAM beneficiaries and those up to and including three years of age would receive an increase of US$60 in their benefit from September to December 2021.Footnote ⁴

The other measure related to cash transfers was the introduction of a new program called Canasta de Emergencia, which applied in supermarkets and other food stores. The amount received with Canasta de Emergencia was around US$30. This transfer was targeted at informal workers not covered by TUS or AFAM or any contributory public program. At its beginning, the program involved the delivery of emergency food baskets, but in May 2022 it was changed to a cash-transfer modality.Footnote ⁵

The benefit could be requested via telephone, with a code then being sent by SMS for redemption. The code was unique to each user and nontransferable, associated with a particular identity card and a cell phone. It is important to note that, given the design of the transfer, any person not registered as a formal worker (or a beneficiary of social security, TUS, and AFAM) could participate in Canasta de Emergencia. Someone living in a household covered by AFAM or TUS could also receive the transfer if he or she was not the benefit holder. In fact, according to the household survey, of all households receiving at least one Canasta de Emergencia, 40 percent in 2020 and 32 percent in 2021 also received TUS or AFAM (see Table A1 in the appendix). Unfortunately, administrative data related to this program, which would have allowed us to understand the profile of beneficiaries and the overlapping of policies, are not available, so we can explore these aspects only through household surveys.

Both AFAM and TUS beneficiaries are concentrated in the lower income deciles, and both transfers present good targeting performance (Figure 3; see also Figure A2).Footnote ⁶ The available information also indicates that the increase in the coverage of food baskets was in the lower part of the distribution, although leakages were also detected (Figure 3). Whereas in 2019 only 0.2 percent of households received a food basket, the coverage increased to 3.1 percent in 2020 and 6.5 percent in 2021. In terms of the amounts of transfers, TUS was on average higher in 2020 than AFAM, but that reversed in 2021, when transfers of existing programs were channeled through AFAM (and not through AFAM or TUS as during 2020) and reinforced by the pilot program for Bono Crianza during the last months of 2021.

Figure 3.

Based on Uruguayan household surveys (ECH). Coverage and average amounts of cash-transfer programs, by income deciles.

Notes: AFAM = Asignaciones Familiares (conditional cash transfer for families with children); TUS = Tarjeta Uruguay Social (unconditional cash transfer for extremely vulnerable households); food baskets = Canasta de Emergencia (emergency food assistance for informal workers). Amounts in constant 2019 Uruguayan pesos. Deciles based on per capita disposable income including all transfers.

Concerning unemployment insurance, this was an effective tool for containing the effects of the COVID-19 crisis in Uruguay (Amarante et al. Reference Amarante, Bucheli and Scalese2021; Marinakis Reference Marinakis2020). Regulatory modifications were implemented to facilitate access to the program (less demanding requirements), and in April 2020, the number of entrants to the program was eight times higher than during the previous two months (Figure 4). Most of the admissions to unemployment insurance after the crisis caused by COVID-19 occurred in the modality of suspension, which acts as a subsidy for temporary reductions in the number of hours worked in firms that face adverse demand shocks, to avoid layoffs. In 2021, the relative importance of the different causes changed, with that of reduction gaining importance.

Figure 4.

Monthly new entrants to the unemployment insurance program, by type.

Notes: Based on Anuario Estadístico, Banco de Previsión Social. Layoff = traditional unemployment insurance for dismissed workers; suspension = temporary suspension of work (job preserved, partial income via unemployment insurance [UI]); reduction = reduced working hours or days (wage top-up via UI).

The percentage of households benefiting from the unemployment insurance program increased systematically in 2020 and 2021 over 2019, and it also changed its profile. The COVID-19 unemployment insurance had a higher incidence in the lower deciles, and it consistently resulted in lower average benefits than those of regular unemployment insurance (see Table A2 in the appendix). The flexibility of the program seems to have facilitated the incorporation of more vulnerable individuals into the pool of formal workers. In any case, beneficiaries of unemployment insurance are much more spread out along the income distribution than are beneficiaries of cash transfers (Figure A2).

In terms of resources, the flexibility of unemployment insurance accounted for 33 percent of the total amount that the government assigned to fight COVID-19 in 2020 (via the Coronavirus Fund), whereas cash transfers represented 18 percent of this fund during the same year (10 percent corresponding to Canasta de Emergencia, 4 percent to income supplements via TUS, and 4 percent to income supplements via AFAM; see Table A3).Footnote ⁷ It is important to note that, whereas cash-transfer programs (especially TUS and to a lesser extent AFAM) are targeted at vulnerable households, the unemployment insurance program covers private formal workers, who generally do not belong to the lower income deciles.

On the tax side, a temporary income tax (COVID-19 Emergency Tax) for public workers (excluding health workers) and pensioners receiving income above a certain threshold was introduced. This tax was levied for two months in 2020 and two months in 2021. For those receiving a gross monthly income over $U120,000 (around US$3,000), four income brackets were defined, and a mean tax rate (increasing with income) was applied. De Rosa et al. (Reference De Rosa, Lanzilotta, Perazzo and Vigorito2020) indicate that those who paid the emergency tax belonged to the upper 4 percent of the distribution of personal incomes. However, around 75 percent of those belonging to this upper 4 percent did not have to contribute to the tax (including private workers and employers), which poses serious doubts about the horizontal equity of the policy design. Moreover, although these taxpayers are among the highest-income groups in the distribution of income earners, their households are located along the entire income distribution and are not concentrated in the higher portion of the distribution of household income. As De Rosa et al. (Reference De Rosa, Lanzilotta, Perazzo and Vigorito2020) emphasize, an alternative way to identify the contributory capacity of income earners would be to consider their total income instead of relying exclusively on the institutional sector of employment. This could be done, using the mechanisms already in place, through personal income tax (IRPF) and include both capital and labor incomes.

In fact, all payers of the COVID-19 Emergency Tax also paid the main direct taxes in Uruguay: the Impuesto a la Renta de las Personas Físicas (IRPF) and the Impuesto a la Asistencia de la Seguridad Social (IASS) on pensions. The IRPF is a dual tax on residents’ labor and capital income.Footnote ⁸

Data

Our analysis is based on representative household survey data from the National Household Survey of Uruguay, collected by INE, for the years 2019, 2020, and 2021.Footnote ⁹ The survey is nationally representative and contains detailed information about personal and household characteristics, as well as detailed data on labor and nonlabor income, contributory and noncontributory pensions, and cash transfers, among other things. Both pension and labor incomes are collected in net terms (after direct taxes). As in many countries, the COVID-19 outbreak affected survey data collection and in-person surveys were replaced by phone surveys from March 2020 to July 2021. This implied a significant reduction in the length of the questionnaire, but the survey was still able to provide adequate information about the labor market situation and income during the whole period.Footnote ¹⁰ At the same time, methodological changes were implemented, as the survey moved from a cross-sectional design to a rotating panel, but this does not affect cross-sectional comparisons (see INE 2022).

An important aspect of the household survey is its ability to capture the payment of extra conditional cash transfers associated with COVID-19. The household survey identifies the beneficiaries of cash transfers, both TUS and AFAM, and the amounts received via each.Footnote ¹¹ The analysis of the amounts collected in the survey shows no variation in 2020 and 2021 with respect to 2019. This may be because families were not able to identify the increase in transfers, which was announced month by month and was not constant over time. It may also be due to problems in the quality of the survey data. If we impute the income corresponding to the transfers in each month during the three years, according to the design rules of the programs and the COVID-19 supplements, there is a significant difference from the amounts reported by the survey (Figure 5). During 2019 and in months when no COVID-19 complements were paid, the imputed amounts coincide with the figures collected by the household survey. With this analysis, and given that our main objective is to analyze the impacts on poverty and inequality of the measures implemented to address COVID-19, including cash transfers, we opted to construct a new vector of household income. This vector is essentially the same as the one reported in the household survey, but instead of taking the transfer amounts collected in the survey, which in our view does not appropriately reflect the COVID-19 supplements, it considers the imputed TUS and AFAM amounts. This is the baseline household income in our analysis.

Figure 5.

Comparison of reported vs. imputed average cash transfers.

Notes: Based on Uruguayan household surveys (ECH). Reported amounts from household survey responses. Imputed amounts calculated based on program rules and COVID-19 supplements. Includes TUS (Tarjeta Uruguay Social) and AFAM (Asignaciones Familiares) transfers. Amounts in constant Uruguayan pesos.

In the case of the emergency food baskets, however, we consider the values collected by the household survey. It is important to note that a household may receive more than one of these baskets, due to the design of the program.

As mentioned, the Uruguayan household survey does not have information about direct taxes and contributions. To analyze the effects of the COVID Emergency Tax, which is levied on nominal remuneration and pensions, we estimate the contributions paid by employers and employees, as well as pensioners, to reconstruct the taxable income and simulate the COVID Emergency Tax. We assume that workers, who do not contribute to the social security system, do not pay direct taxes.

Method

This research is based on a static microsimulation approach, and so constitutes a partial equilibrium analysis. This study employs an independently developed static microsimulation model, programmed specifically for Uruguay’s tax-benefit system.Footnote ¹² Microsimulation models apply the main rules that define the tax and benefit system to each household in a representative sample of the population (household survey). Besides showing the redistributive impact of the existing systems, these models allow us to assess the redistributive impacts of reforms on taxes and/or benefits. We apply the Uruguayan tax and benefit rules to the Uruguayan household survey for the years 2019, 2020, and 2021.

The modeling of Uruguay’s income tax system required reconstructing gross incomes from the net amounts reported in the household survey. We estimated gross labor income by applying the inverse of the tax and social security contribution schedules to reported net incomes, using iterative calculations to account for the progressive tax structure. For the COVID Emergency Tax, we identified eligible taxpayers based on employment sector (public workers and pensioners with gross monthly income above $U120,000, excluding health workers) and applied the corresponding progressive rates for the specified periods. Given that the COVID Emergency Tax was levied only for specific months (two months in 2020 and two months in 2021), whereas the household survey data represent annual income, we applied a temporal adjustment to accurately reflect the limited duration of this tax. For eligible taxpayers, we calculated the monthly tax liability based on their reported monthly gross income and applied this tax only for the months when the emergency tax was in effect. The annual impact was then calculated by summing the tax payments across the relevant months while leaving the remaining months untaxed. This approach ensures that our estimates reflect the actual temporal implementation of the policy rather than incorrectly applying the tax across the entire year, which would overestimate its distributional impact.

We reconstruct two vectors of income, market income ( $Y_h^{market}$ ) and disposable income ${\rm{\;}}(Y_h^{dis{\rm{\;}}})$ , defined as follows:

(1) $$Y_h^{dis}=Y_h^{market} - \left(\sum\nolimits_i {{T_i}} + \sum\nolimits_i {T_i^{Covid}} \right) + \left( {\;\sum\nolimits_i {{B_i}} + \sum\nolimits_i {B_i^{Covid}} } \right),$$

where $\sum\nolimits_i {{T_i}} $ represents regular direct taxes paid by members i of household h (IASS and IRPF), $\sum\nolimits_i {T_i^{Covid}} $ represents COVID-19-related taxes (COVID Emergency Tax), $\sum\nolimits_i {{B_i}} $ represents regular benefits received by household members (standard cash transfers and unemployment insurance), and $\sum\nolimits_i {B_i^{Covid}} $ represents COVID-related benefit modifications (transfer supplements, emergency food baskets, and COVID-19 unemployment benefits). For 2019, all COVID-19-related components ( $\sum\nolimits_i {T_i^{Covid}} $ and $\sum\nolimits_i {B_i^{Covid}} $ ) equal zero.

To reflect the impact of social benefits on poverty and extreme poverty incidence, we compute the traditional poverty measures based on disposable income and recalculate them assuming that each of the benefits was zero. For the poverty analysis, we use the official national poverty and extreme poverty lines as defined by INE (2006). Uruguay employs household-composition-adjusted poverty lines that vary according to household size, as well as geographic location (urban vs. rural areas), reflecting different consumption needs and cost structures. The poverty line represents the cost of a basic basket of goods and services needed to meet minimum living standards, whereas the extreme poverty (indigence) line represents the cost of a basic food basket required for survival. Expressed in per capita terms, the average poverty line was equivalent to approximately US$427 per month at 2019 purchasing power parity (PPP), whereas the extreme poverty line averaged approximately US$137 per month at 2019 PPP across the analysis period (INE 2006). The COVID Emergency Tax is not considered in the poverty analysis, as it only affects higher-income households.

To analyze the progressivity of each intervention, we display concentration curves. To consider the redistributive impact of the interventions, we calculate the marginal effect of each tax and transfer, by recalculating the Gini index and excluding the analyzed instrument from the income vectors. In this way, we decompose the total redistributive action during 2020 and 2021 into the part that is due to the precrisis tax-benefit program that Uruguay had and the part that is due to COVID-19-related emergency measures.

Changes in household income, poverty and indigence

Between 2019 and 2020, the decline in (per capita) household disposable income was higher, in relative terms, for households at the lower part of the distribution, as shown in Figure 6, which presents changes in household income by income decile. Household income recovered during 2021, and recovery was stronger for households in the lower deciles. Comparing 2019 and 2021, the decline in household income was rather similar along the income distribution, except in the first decile, which presented a lower decline. Therefore, poverty and inequality increased during 2020 but figures improved during 2021, although they did not return to pre-COVID-19 levels (Figure 6 and Table 1).Footnote ¹³ Previous results for Ecuador (Jara et al. Reference Jara, Montesdeoca and Tasseva2021), Colombia (Corredor et al. Reference Corredor, Ríos and Rodríguez2021), and Argentina, Brazil, Colombia, and Mexico (Lustig et al. Reference Lustig, Pabon, Sanz and Younger2023) found that the middle strata experienced the greatest loss of income.

Figure 6.

Changes in per capita household disposable income, by income decile.

Notes: Based on Uruguayan household surveys (ECH). Growth rates calculated as percentage change from previous year. Negative values indicate income decline. Deciles based on per capita disposable income in base year.

Table 1.

Impact of social protection benefits on poverty incidence (all households).

Table 1. Long description

A table comparing the impact of social protection benefits on poverty incidence across different scenarios and years. The table has 16 rows and 6 columns. The columns are labeled Poverty headcount ratio (%) for the years 2019, 2020, and 2021, and Poverty reduction (pp) for the years 2019, 2020, and 2021. The rows are labeled with different scenarios, including a baseline scenario and various counterfactual scenarios without specific benefits. Row 1: Baseline scenario, With all benefits, 8.6, 11.0, 9.5,, , . Row 2: Counterfactual scenarios, Without AFAM (regular), 9.9, 12.7, 10.9, 1.4, 1.6, 1.4. Row 3: Without AFAM COVID, n.a., 11.3, 10.7, n.a., 0.3, 1.1. Row 4: Without AFAM (all), 9.9, 12.9, 12.0, 1.4, 1.9, 2.4. Row 5: Without TUS (regular), 9.2, 11.7, 10.2, 0.7, 0.7, 0.6. Row 6: Without TUS (COVID), n.a., 11.4, n.a., n.a., 0.3, n.a. Row 7: Without TUS (all), 9.2, 12.0, 10.2, 0.7, 1.0, 0.6. Row 8: Without food baskets, n.a., 11.1, 9.8, n.a., 0.1, 0.3. Row 9: Without UI (regular), 9.0, 11.4, 9.7, 0.4, 0.4, 0.2. Row 10: Without UI (COVID), n.a., 11.8, 9.9, n.a., 0.8, 0.4. Row 11: Without UI (all), 9.0, 12.2, 10.1, 0.4, 1.2, 0.6. Row 12: Without regular transfers (AFAM, TUS), 10.6, 13.2, 11.4, 2.0, 2.2, 1.9. Row 13: Without regular and COVID-19 transfers (AFAM all, TUS all, Food baskets), 10.6, 13.8, 12.6, 2.0, 2.8, 3.0. Row 14: Without regular benefits (AFAM, TUS, UI), 11.0, 13.6, 11.6, 2.5, 2.6, 2.1. Row 15: Without regular and COVID-19 benefits (AFAM all, TUS all, Canasta, UI all), 11.0, 14.9, 13.1, 2.5, 3.9, 3.6.

Notes: Based on Uruguayan household surveys (ECH). Shows poverty headcount ratios with and without specific social protection benefits. Regular transfers include standard AFAM (Asignaciones Familiares) and TUS (Tarjeta Uruguay Social) programs. COVID-19 transfers include AFAM and TUS supplements and food baskets (Canasta de Emergencia). Regular benefits include regular transfers plus standard unemployment insurance. COVID-19 benefits include COVID-19 transfers plus extended unemployment insurance. UI stands for Unemployment Insurance. Each row excludes only the indicated benefit while including all other benefits in the calculation. N.a. = program not implemented in that year. Poverty reduction calculated as the difference between each counterfactual scenario (without the specified program) and the baseline scenario (with all programs).

Both in 2020 and 2021, social protection benefits alleviated pressure on household budgets, reinforcing the role of traditional cash transfers. Figure 7 presents the composition of household income separating pre-COVID-19 benefits and benefits implemented due to COVID-19. The powerful role of the preexisting benefits in the lower part of the distribution, and the weak reinforcement through COVID-19 measures, is depicted.

Figure 7.

Social protection benefits as share of per capita household disposable income, by income percentiles.

Notes: Based on Uruguayan household surveys (ECH). Figure shows social protection benefits as a percentage of total household disposable income across the income distribution. Pre-COVID-19 benefits include regular AFAM and TUS transfers plus standard unemployment insurance. COVID-19 benefits include transfer supplements, emergency food baskets (Canasta de Emergencia), and expanded unemployment insurance. Figure does not show wages, pensions, or capital income components.

To reflect the impact of the different measures adopted, we depart from poverty and indigence incidence, measured by disposable household income, and recalculate poverty and indigence rates deducting benefits (cash transfers, food baskets, and unemployment insurance) one by one. We consider the separate effects of the existing programs and the modifications due to COVID-19. Table 1 and Figure A3 in the appendix show the main results of this analysis for all households.

If we consider the regular components of the benefit system, the main effect on poverty is due to AFAM. The regular transfer allows for the poverty incidence to decrease by around 1.6 pp. The supplements imply an additional 0.3-pp deduction in 2020, climbing to 1.1 in 2021 when the design of the supplement was changed and an additional supplement for children aged zero to three was given, as discussed in the previous section.

The effect of the regular TUS transfer on poverty incidence is lower, as it is targeted at more vulnerable households, whose income is furthest from the poverty threshold. In this case, the effect is 0.7 pp in 2020 and 0.6 in 2021, with an additional 0.3 pp in 2020 thanks to the supplemental value, which was not implemented in 2021. An even smaller decline than the one obtained by the supplemental TUS is achieved through the food basket, which implies a 0.1 and 0.3 pp decline in poverty in 2020 and 2021, respectively, underlining the negligible effects of this COVID-19 transfer due to the low amounts involved. The effect of the traditional unemployment insurance payment is around 0.4 pp in 2020, but the modifications to this program due to COVID-19 significantly boosted the poverty impact of this benefit, suggesting a reduction of an additional 0.8 and 0.4 pp in 2020 and 2021. This means that, considering the benefits globally, the major impact on poverty incidence was obtained through AFAM, although the flexibility of unemployment insurance had a relevant effect, especially in 2020.

In summary, without any social protection, poverty would have reached 14.9 percent in 2020 and 13.1 percent in 2021, compared to the observed rates of 11.0 percent and 9.5 percent with the full system in place. Preexisting social protection programs (regular transfers and unemployment insurance) accounted for reductions of 2.6 and 2.1 pp in 2020 and 2021, respectively. The modifications introduced to face COVID-19 provided an additional reduction of 1.3 pp in 2020 and 1.5 pp in 2021, demonstrating how emergency measures complemented the existing social protection framework during the crisis. These results are consistent with Rodríguez et al. (Reference Rodríguez, Jara, Dondo, Arancibia, Macas Romero, Riella, Urraburu, Jouste, Kanbur, Pirttilä and Rattenhuber2025), who found that for Uruguay, COVID-19 emergency policies provided cushioning effects primarily for households at the bottom of the distribution during 2020, though our extended analysis through 2021 reveals the sustained and evolving impact of these measures.

If we restrict the analysis to households with children (Table A4 and Figure A4 in the Appendix), the effect of AFAM is higher, especially in 2021. It is important to remember that poverty in Uruguay has a clearly defined profile, with a high concentration of households with children (Colacce and Tenenbaum Reference Colacce and Tenenbaum2016; De Armas Reference De Armas2017). Again, AFAM is the main COVID-19 measure in terms of impact on poverty, but the unemployment benefits also played a role for these households. If no modifications had been undertaken in social protection as a response to COVID-19, poverty would have been 1.8 and 2.4 pp higher among households with children in 2020 and 2021, respectively. In comparative terms, this effect with COVID-19 measures is an intermediate result when compared with regional results (see Figure 2) and a meager one when compared with results in the developed world.

Uruguay’s extremely low indigence rates throughout the study period reflect the country’s relatively developed social protection system compared to regional standards. While Table A5 and Figure A5 present the technical impact of different benefits on extreme poverty, the near-zero baseline levels limit the analytical significance of these comparisons. The fact that indigence remained minimal even during the COVID-19 crisis demonstrates the effectiveness of Uruguay’s existing social safety net in preventing the most severe forms of deprivation. Although all programs showed some effect on extreme poverty, results should be interpreted cautiously, given the very small base rates from which changes are calculated.

The global effects of the additional COVID-19 benefits were moderate for all households in terms of poverty reduction and had a slighter higher impact when we restrict the analysis to households with children (Figure 8).

Figure 8.

Poverty reduction effect of social protection benefits.

Notes: Based on Uruguayan household surveys (ECH). Shows the reduction in poverty incidence when social protection benefits are included versus excluded from household income. Regular benefits include preexisting cash-transfer programs and unemployment insurance. COVID-19 benefits include pandemic-related transfer supplements, emergency food assistance, and expanded unemployment insurance programs implemented during the crisis.

Progressivity of benefits and taxes and redistributive impact

The redistributive impact of each instrument will depend on the progressive character of the instrument, population coverage, and the amount involved. As Figure 9 shows, all the benefits that we are considering are progressive, as their concentration curves lie above the pretransfer Lorenz curve, indicating that benefits as a share of pretransfer income decline with income. Moreover, all transfers (TUS, AFAM, and the food baskets) are pro-poor, as their concentration curves lie above the diagonal, TUS being the most pro-poor transfer and food baskets the least. This means that the benefit declines with pretransfer income in absolute terms (and not only as a share). The unemployment insurance benefit, in turn, is progressive but not pro-poor.

Figure 9.

Concentration curves for social protection benefits.

Notes: Based on Uruguayan household surveys (ECH). Benefits are progressive when the concentration curve lies above the pretransfer Lorenz curve. Benefits are pro-poor when concentration curve lies above the 45-degree line (diagonal). AFAM = Asignaciones Familiares; TUS = Tarjeta Uruguay Social; UI = Unemployment Insurance; Population ranked by household disposable income excluding the analyzed benefit.

Figure 9. Long description

The image contains four concentration curves, each representing different social protection benefits for households with children in 2020 and 2021. Panel A shows the concentration curves for 2020 transfers, while Panel B shows the concentration curves for 2021 transfers. Panel C displays the concentration curves for 2020 unemployment insurance benefits, and Panel D shows the concentration curves for 2021 unemployment insurance benefits. Each panel includes multiple lines representing different types of benefits: household income without benefits, regular AFAM, COVID AFAM, regular TUS, COVID TUS, and food baskets for Panels A and B; household income without benefits, regular unemployment insurance, and COVID unemployment insurance for Panels C and D. The x-axis in each panel represents the cumulative percentage of households, ranging from 1 to 100 percent. The y-axis represents the cumulative percentage of benefits received, also ranging from 0 to 100 percent. The red diagonal line in each panel represents the line of equality, where each decile of the population receives an equal share of the benefits. The curves above the red line indicate that the benefits are concentrated among the poorer segments of the population. The curves for regular AFAM, COVID AFAM, regular TUS, and COVID TUS in Panels A and B show that these benefits are more concentrated among the poorer households compared to household income without benefits. Similarly, the curves for regular unemployment insurance and COVID unemployment insurance in Panels C and D indicate that these benefits are also more concentrated among the poorer households. The food baskets in Panel A and B show a similar pattern of concentration among the poorer households. The trends in the curves suggest that the social protection measures implemented during the COVID-19 pandemic had a significant impact on reducing poverty among households with children.

Concerning taxes, all direct taxes are globally progressive, as the concentration curve lies below the pretax Lorenz curve, and the COVID Emergency Tax is more progressive (Figure 10).

Figure 10.

Concentration curves for direct taxes.

Notes: Based on Uruguayan household surveys (ECH). Taxes are progressive when the concentration curve lies below the pretax Lorenz curve. IRPF = personal income tax; IASS = social security contribution tax on pensions. COVID Emergency Tax applied to high-income public workers and pensioners for limited periods. Population ranked by household income before taxes.

To measure the redistributive impact of each instrument, we calculate its marginal contribution to inequality reduction. This marginal contribution is the difference between the Gini index calculated on household disposable income with all benefits and taxes included, and the Gini index calculated excluding the specific benefit (or including the tax) that we are analyzing. Direct taxes have a significant redistributive effect, leading to decreases in inequality of around 3 pp, mainly through IRPF and IASS. The COVID Emergency Tax does not change income distribution indicators, primarily because it was applied to a small group of workers and for very limited periods.

The higher marginal contribution is given by the action of AFAM, which in regular times implies a reduction of around half a percentage point in the Gini index. During COVID-19 this effect climbed to 0.6 in 2020 and 0.9 in 2021 (see Table 2), when the pilot program for smaller children was implemented. TUS also had a nearly half-point impact on the Gini index in 2020, when the supplemental transfer was introduced. The marginal contribution of the unemployment insurance program was doubled thanks to the COVID-19 modifications. During 2020 it implied an equalizing effect of half a percentage point on the Gini index. Taken together, the modifications to the social protection system introduced during the COVID-19 crisis prevented the Gini index from increasing by an additional 0.6–0.7 percentage points.

Table 2.

Redistributive impact of fiscal instruments on income inequality.

Table 2. Long description

A table comparing the Gini index and Gini reduction across different scenarios and years. The table has 16 rows and 6 columns. The columns are labeled as follows: Gini index for the years 2019, 2020, and 2021, and Gini reduction for the years 2019, 2020, and 2021. The rows are labeled with different scenarios, including a baseline scenario and various counterfactual scenarios without specific benefits or transfers. Row 1: Baseline scenario, With all benefits, Gini index: 0.341, 0.350, 0.349, Gini reduction: empty, empty, empty. Row 2: Counterfactual scenarios, Without AFAM (regular), Gini index: 0.346, 0.355, 0.354, Gini reduction: 0.005, 0.005, 0.005. Row 3: Without AFAM COVID, Gini index: n.a., 0.351, 0.352, Gini reduction: n.a., 0.001, 0.003. Row 4: Without AFAM (all), Gini index: 0.346, 0.356, 0.358, Gini reduction: 0.005, 0.006, 0.009. Row 5: Without TUS (regular), Gini index: 0.344, 0.353, 0.352, Gini reduction: 0.003, 0.003, 0.003. Row 6: Without TUS (COVID), Gini index: n.a., 0.351, n.a., Gini reduction: n.a., 0.001, n.a. Row 7: Without TUS (all), Gini index: 0.344, 0.354, 0.352, Gini reduction: 0.003, 0.004, 0.003. Row 8: Without Food Baskets, Gini index: n.a., 0.350, 0.350, Gini reduction: n.a., 0.000, 0.001. Row 9: Without UI (regular), Gini index: 0.343, 0.351, 0.350, Gini reduction: 0.002, 0.001, 0.001. Row 10: Without UI (COVID), Gini index: n.a., 0.353, 0.350, Gini reduction: n.a., 0.003, 0.001. Row 11: Without UI (all), Gini index: 0.343, 0.355, 0.351, Gini reduction: 0.002, 0.005, 0.002. Row 12: Without regular transfers (AFAM, TUS), Gini index: 0.349, 0.359, 0.357, Gini reduction: 0.008, 0.009, 0.008. Row 13: Without regular and COVID-19 transfers (AFAM all, TUS all, Food baskets), Gini index: 0.349, 0.362, 0.362, Gini reduction: 0.008, 0.012, 0.013. Row 14: Without regular benefits (AFAM, TUS, UI), Gini index: 0.351, 0.361, 0.358, Gini reduction: 0.010, 0.011, 0.009. Row 15: Without regular and COVID-19 benefits (AFAM all, TUS all, Canasta, UI all), Gini index: 0.351, 0.367, 0.365, Gini reduction: 0.010, 0.017, 0.016.

Notes: based on Uruguayan household surveys (ECH). Shows Gini index values with and without specific fiscal instruments. Regular transfers include standard AFAM (Asignaciones Familiares) and TUS (Tarjeta Uruguay Social) programs. COVID-19 transfers include AFAM and TUS supplements and food baskets (Canasta de Emergencia). Regular benefits include regular transfers plus standard unemployment insurance. COVID-19 benefits include COVID-19 transfers plus extended unemployment insurance. UI = Unemployment Insurance. Each row excludes only the indicated benefit while including all other benefits in the calculation.