Resumo Estimamos um VAR estrutural (SVAR) baseado em Blanchard e Perotti (2002) com dados atualizados do governo central de Gobetti e Orair (2017) para estimar multiplicadores fiscais. O resultado sugere um multiplicador maior na amostra completa (1997-2018) em relação à amostra pré-crise (até 2014) para benefícios sociais e investimentos públicos. Nosso primeiro exercício avalia a reorientação da política fiscal para os subsídios entre 2011 e 2014, com base nos multiplicadores estimados a partir do SVAR. O PIB no cenário em que os subsídios são destinados aos investimentos públicos estaria 2.7% acima do PIB verdadeiro em 2014. O segundo exercício projeta que, caso os investimentos públicos mantivessem a mesma taxa média de crescimento de 2006-2010 durante a crise, o produto estaria 6% acima do PIB observado em 2017. Além disso, o PIB estaria 2.53% abaixo do verdadeiro se os benefícios sociais não tivessem crescido em 2016 e 2017.

Abstract We estimated a structural VAR based on Blanchard and Perotti (2002)’s methodology with updated data of the central government from Gobetti and Orair (2017) in order to estimate fiscal multipliers. The result suggests a higher multiplier in the complete sample (1997-2018) than in the pre-crisis sample (until 2014) for social benefits and public investments. Our first exercise assesses the reorientation of fiscal policy towards subsidies between 2011 and 2014. GDP in the scenario in which subsidies are destinated to public investments would be 2.7% above the actual GDP in 2014. The second exercise projects that, if public investments had kept the same average growth rate of 2006-2010 during the crisis, GDP would be 6% above the observed GDP in 2017. Also, GDP would be 2.53% below the actual if social benefits had not grown in 2016 and 2017.

Abstract The paper presents empirical evidence from Brazil on the bidirectional relationship between inequality and the Covid-19 crisis: Racial, regional and income inequalities have both aggravated the pandemic and been exacerbated by it. The paper provides a measure of social risk factors to Covid-19, which is positively associated with the number of Covid-19 cases across Brazilian states. This is complemented by a discussion of the country's inequalities in access to healthcare and in the incidence of comorbidities, which elucidates observed disparities in the need of a ventilator and in the number of deaths by Covid-19. Finally, our empirical analysis suggests that the substantive fiscal effort in the country and particularly the implementation of the cash relief program Auxílio Emergencial neutralized the short-term increase in wage disparities but has been insufficient to neutralize the cost of multi-dimensional structural inequalities on Brazil’s death tolls.

Resumen Este artículo presenta evidencia empírica de Brasil sobre la relación bidireccional entre desigualdad y la crisis de Covid-19: las desigualdades raciales, regionales y de ingresos han agravado y se han exacerbado a causa de la pandemia. El estudio construye una medida de factores de riesgo social que se asocia positivamente con el número de casos de Covid-19 en los estados brasileños. Esto se complementa con una discusión sobre las desigualdades en el acceso a salud y en la incidencia de comorbilidades que explica las disparidades observadas en la necesidad de un ventilador durante la pandemia y en el número de muertes. Finalmente, nuestro análisis empírico sugiere que el sustancial esfuerzo fiscal y en particular la implementación del programa Auxílio Emergencial neutralizaron el aumento a corto plazo de las disparidades salariales, pero han sido insuficientes para neutralizar el costo de las desigualdades multidimensionales en el número de muertes en Brasil.

ABSTRACT Two studies were carried out to evaluate the effects of levels of soybean meal produced from damaged soybean grains replacing normal soybean meal on in vitro and in situ digestibility of ruminant diets as well as on intake, digestibility, and animal performance of growing/finishing lambs. In trial 1, we evaluated the in vitro digestibility of each soybean meals (normal and damaged), as well as diets containing levels of the damaged replacing the normal (0, 333, 667 and 1,000 g.kg−1) soybean meal on kinetic parameters of in vitro cumulative gas production, in vitro dry matter and crude protein digestibility, in situ rumen-degradable protein, rumen-undegradable protein, and in vitro intestinal digestibility of rumen-undegradable protein. In trial 2, we used 48 growing/finishing lambs to evaluate the effects of damaged soybean meal levels (0, 333, 667, and 1,000 g.kg−1) replacing normal soybean meal in feedlot diets (20:80 roughage:concentrate ratio) on intake and digestibility of nutrients and on animal performance. The damaged soybean meal presented lower values for total gas production and in vitro dry matter digestibility than normal soybean meal. Higher rumen-undegradable protein was estimated for damaged soybean meal than for the normal and consequently lower rumen-degradable protein for damaged compared to normal. Because of the lower rumen-degradable protein, damaged soybean meal promoted lower in vitro ammonium nitrogen (NH3-N) concentrations than the normal in feedlot diets. In the in vivo trial, there were no effects of damaged soybean meal levels in the diets on intake and digestibility of nutrients (dry matter, organic matter, crude protein, and fiber) as well as on total weight gain, average daily gain, carcass yield, or feeding efficiency. Thus, damaged soybean meal can fully replace the normal one in lamb feedlot diets (in up to of 1,000 g.kg−1 of the normal soybean meal) without causing adverse effects on intake and digestibility of nutrients and on animal performance.