The Covid-19 crisis reduced household income and significantly aggravated poverty and inequality. It led to an overall reduction of income in Portugal, as revealed by the leftward shift of the income distribution (more individuals near or below the poverty line under the crisis scenario, graph 1). More than 400,000 individuals became poor as a consequence of the Covid-19 crisis and the gap between rich and poor (measured by the ratio between the 95th and the 5th income percentiles) increased by about 9%, from 7.6 to 8.3.

The Covid-19 pandemic disrupted economic activity. This research aims to assess the impact of the crisis on poverty and inequality in Portugal and the cushioning effects of the policies enacted. Using the 2019 Inquérito às Condições de Vida e Rendimento (ICOR) to extrapolate 2020 incomes, various macroeconomic scenarios were built to compare (1) crisis versus no-crisis over the year and (2) crisis with and without protective policies over an 8-week full lockdown. For the crisis scenarios, occupation vulnerability to lockdown restrictions was assessed based on whether it was possible to work remotely and whether the occupation was essential. Under the crisis scenarios, non-vulnerable workers kept their full no-crisis income, while the income of vulnerable workers (who were randomly selected by sector to match aggregate statistics on cushioning policies) was reduced. This is true even taking into account the cushioning measures implemented by the Portuguese government, specifically the simplified lay-off regime, targeting dependent workers, and the extraordinary supports for self-employed workers.
    
The results show that the pandemic led to a staggering 25% increase in poverty over a year, when compared to the no-crisis scenario, jeopardising the progress made over the last twenty years and the continuous decrease since 2015, when the poverty rate was 19% (graph 2). As for inequality, the analysis points to a 9% increase (graph 5). Losses are concentrated in some groups, such as the lower- and middle-income groups, the Algarve region and people with only a 9th grade education. The policies enacted mitigated the rise of poverty and inequality: without them, the initial 8-week lockdown alone would have led to a similar impact on poverty and on inequality as simulated for the entire year.

The Covid-19 crisis significantly impacted poverty in Portugal. It is estimated that more than 400,000 people fell below the poverty line (which was maintained from the no-crisis scenario to avoid an artificially small increase due to the decline of the median income). This represents an increase in the AROP rate of 25%, from 18.3% (in the baseline scenario) to 23% (in the crisis scenario), inverting the downward trend started in 2015 (graph 2). The poor are also poorer: poverty intensity, which captures how far the poor are from the poverty line, is estimated to have increased by 8%, compared to the no-crisis scenario.

Nevertheless, the policies enacted by the government mitigated the increase in poverty. Without these policies, an 8-week lockdown (in which those who cannot work remotely and are not considered essential stop earning) would have led to an increase in the poverty rate of over 20% and an increase in the poverty intensity of nearly 8% (graph 3) compared to the no-crisis scenario, values similar to estimates for an entire year. In the scenario of an 8-week lockdown that includes the simplified lay-off regime for dependent workers and the extraordinary supports to the self-employed, both poverty measures would have only increased by around 4%.

Those who could work remotely or who worked in activities deemed essential kept their salaries, while those who could not work suffered significant losses. The likelihood of being part of this second group is higher among those in the lower income groups. As a result, the share of income lost due to the crisis was greater for those already at the lower end of the income distribution, increasing the gap between rich and poor (graph 4). In the no-crisis baseline scenario, the income of those at the 95th percentile was 7.6 times larger than the income of those at the 5th percentile, whereas with the Covid-19 crisis, it jumps to 8.3 times, that is, a 9% increase in inequality (graph 5).

Policy interventions were important to prevent a more substantial rise in inequality. In an 8-week lockdown, if those who could not work had been left without any income from paid work, the ratio between the 95th and 5th percentiles would have increased by 8% compared to the no-crisis scenario, about the same as simulated for an entire year. With interventions, the ratio between the 95th and 5th percentiles is unaffected over an 8-week lockdown.

According to these simulations, people lost on average 7% of their annual income. This number, however, hides significant variability across income groups. While all income groups suffered with the crisis, the negative impact was considerably skewed towards the bottom half of the no-crisis income distribution (graph 6). Furthermore, looking solely at those with income losses, the average loss rises to nearly 16%, with the lower- and lower-middle income groups suffering the most (19% and 16%, respectively) (graph 6).

The policies enacted, nevertheless, evened the income losses across groups. Without these measures, the first 8-week lockdown alone would have been enough to cause a 7% income loss for people with incomes below the median. Looking at the same period with protective policies, the income loss is smoother: the average, for all income groups, is less than 2%.

However, the asymmetric nature of the crisis is not restricted to original incomes. The simulation suggests that people from Algarve lost, on average, 9% of their yearly income, and people with only a 9th grade education lost, on average, over 10% of their yearly income.

While this study focused on the divide between workers who can work remotely or are essential (non-vulnerable), and those who cannot (vulnerable), this crisis had very complex outcomes, some of which are not addressed by this study. First, since it was only possible to access the evolution of the total number of workers who benefited from the support programmes, workers were assigned randomly to these programmes, limiting the accuracy of the simulation. Second, in the simulations, only labour income was affected, disregarding eventual changes in non-labour income such as income from investments or state-provided income for low-income families. Third, for simplicity, the analysis assumes a constant rate of employment, while the crisis likely increased unemployment. To minimise estimation errors deriving from this assumption, the effect of the protective policies was assessed only over an 8-week full lockdown. Lastly, for the sake of conciseness, only the AROP rate and the ratio between the 95th and 5th income percentiles are presented as the main poverty and inequality indicators, respectively. The results are also robust to the use of other measures, but these two measures alone are detailed in this article as they reflect the impact of the crisis while also being easily understood.

The Covid-19 pandemic is inevitably linked to the social and economic crisis it triggered. This study estimates that, as a result of the pandemic, the poverty rate in Portugal has increased by a staggering 25% compared to the no-crisis scenario, and the gap between rich and poor has greatly widened as a result of the asymmetric nature of this crisis, with a 9% increase of inequality as measured by the ratio between the 95th and 5th income percentiles, over a year.

Numerous policies were enacted to mitigate the impact of the crisis. This study shows that the simplified lay-off regime for dependent workers and the extraordinary support to the self-employed were effective: without them, the initial 8-week lockdown would have led to a similar impact on poverty and inequality as simulated for the entire year.

The ongoing pandemic and the resulting economic crisis may cause substantial fiscal challenges, as government efforts of the same magnitude may be difficult to sustain for a prolonged period. In light of these results, it is clear that without a strong recovery, a reduction in support could backfire and have drastic consequences for poverty and inequality.

Decree-Law n. º 12-A/2020 from April 6th. Diário da República nº 68/2020, 3º Suplemento, Série I. Presidency of the Council of Ministers. 
Espiga, F. (2020). Wage inequality: Trends and Drivers in Portugal. Master dissertation in Economics. Católica-Lisbon School of Business and Economics, Universidade Católica Portuguesa.
Eurostat. (2018, August 1st). Glossary: At-risk-of-poverty-rate. Retrieved from https://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:At-risk-of-poverty_rate.
Foschiatti, C. B. & Gasparini, L. (2020). El Impacto Asimétrico de la Cuarentena. Documentos de Trabajo del CEDLAS Nº 261, April, 2020, CEDLAS-Universidad Nacional de La Plata.
Fundação Francisco Manuel dos Santos. (2021, February 19th). Taxa de risco de pobreza: antes e após transferências sociais. PORDATA. https://www.pordata.pt/Portugal/Taxa+de+risco+de+pobreza+antes+e+ap%C3%B3s+transfer%C3%AAncias+sociais-2399
Instituto Nacional de Estatística. (2019). Inquérito às Condições de Vida e Rendimento.
Ministério do Trabalho e Segurança Social. (n.d.). Indicadores Covid-19 MTSSS - Gabinete de Estratégia e Planeamento. Gabinete de Estratégia e Planeamento (Ministério do Trabalho, Solidariedade e Segurança Social). http://www.gep.mtsss.gov.pt/indicadores-covid-19-mtsss
Neri, M. (2020, August). What are the immediate effects of the Covid-19 pandemic on the Brazilian Economic Classes? FGV SOCIAL CENTRO DE POLÍTICAS SOCIAIS. https://fgveurope.fgv.br/sites/fgveurope.fgv.br/files/downloads/covideasclasseseconomicas_fgv_social_neri-eng.pdf
Ordinance n.º 71-A/2020 from March 15th. Diário da República n.º 52-A/2020, 1º Suplemento, Série I. Ministry of Employment, Solidarity and Social Security. Lisbon.
Silva, J., Gaspar, M., Fernandes, P., Kouhen, K. & Leitão, M. (2020). Who will bear the costs of the  COVID-19  crisis?  Analysis of  jobs at  risk in  Portugal,  Universidade Católica Portuguesa.