Efficiency of Social Expenditure Levels in Reducing Poverty Risk in the EU‐28

We evaluate the efficiency of social expenditure (SE) to reduce poverty in the European Union (EU) between 2007 and 2015. The data are programs provided by the EU‐SILC and the ESSPROS. Incorporating efficiency analysis improves the results of the analysis of that complex relationship. We found that the correlation between SE and the levels of poverty (and the correlation between SE and the capacity to reduce poverty) is not strong. In addition, the correlation between SE and the levels of poverty before SE (and the correlation between SE and the levels of inequality before SE) is not significant. Our analysis of the efficiency identifies four groups of countries. The first group has a correspondence with the Continental and Nordic Welfare models (Esping‐Andersen, 1990). They have a great level of SE, and their efficiency is above the EU‐28 average. The second group presents the highest efficiency values. The third group is formed by the Mediterranean welfare states, which present a moderate level of expenditure but have low efficiency. A fourth group composed of non‐EU‐15 countries presents a low SE with under/on average efficiency. We also observe that SEs in family/children, housing, and sickness/healthcare are significantly and positively correlated with efficiency measures. Likewise, in the EU‐15, social exclusion expense exhibits a great positive relationship with the efficiency in poverty reduction. 我们评估了2007年至2015年间社会支出(SE)在减少欧盟国家贫困时的效率。数据由欧盟收入与生活条件统计资料(EU‐SILC) 和欧洲综合社会保护统计体系(the ESSPROS)等项目提供。通过结合效率分析, 提高了该复杂关系的分析结果。我们发现, 社会支出的程度与贫困程度、减贫能力并不存在清晰关系, 同时其与贫困程度或不平等程度的关系在社会支出出现之前并不显著。我们的效率分析识别了四组国家。第一组与洲际和北欧福利模式(Esping‐Andersen, 1990)存在联系。这组国家的社会支出程度相当高, 并且社会支出效率超过欧盟28国的平均水平。第二组展现的效率值最高。第三组由地中海福利国组成, 展现了中等程度的支出, 但支出效率低。第四组由非欧盟15国组成, 其展现的社会支出少, 支出效率等于或低于平均水平。我们还发现, 用于家庭/儿童、住房与疾病/医疗的社会支出与效率措施显著相关, 且为正相关。同样, 欧盟15国中用于社会排斥的支出与减贫效率呈显著正相关, 养老金与减贫效率呈显著负相关。 We evaluate the efficiency of social expenditure (SE) to reduce poverty in the European Union between 2007 and 2015. The data are provided by the EU‐SILC and the ESSPROS programmes. Incorporating efficiency analysis improves the results of the analysis of that complex relationship. We found that the level of SE has not a clear correlation with levels of poverty and with their capacity to reduce poverty, while its correlation with levels of poverty or inequality before SE is not significant. Our analysis of the efficiency identifies four groups of countries. The first group has a correspondence with the Continental and Nordic Welfare models (Esping‐Andersen, 1990). They have a great level of SE and their efficiency is above UE28 average. The second group presents the highest efficiency values. The third group is formed by the Mediterranean welfare states, which present a moderate level of expenditure, but have low efficiency. A fourth group composed by non‐EU‐15 countries presents a low SE with under/on average efficiency. We also observe that SE in family/children, housing and sickness/healthcare are significantly and positively correlated with efficiency measures. Likewise, in UE15 also social exclusion (pensions) expense exhibit a great positive (negative) relationship with the efficiency in poverty reduction.