The Welfare of Syrian Refugees: Evidence from Jordan and Lebanon by Paolo Verme; Chiara Gigliarano; Christina Wieser; Kerren Hedlund; Marc Petzoldt; Marco Santacroce

Author:Paolo Verme; Chiara Gigliarano; Christina Wieser; Kerren Hedlund; Marc Petzoldt; Marco Santacroce
Language: eng
Format: epub

Source: Estimations based on JD-HV2 data.

Table 2.15 Welfare Model

Source: Estimations based on JD-HV2 data.

The first and most important variable of the model is case size. This is partly by construction, as the dependent variable is in per capita terms measured with case size on the denominator. However, case size can potentially take any sign and is used in combination with all other variables in the model, which means that the coefficient is not predictable by construction. Removing this variable alone would reduce the explanatory power of the model by 18.1 percent. The variable is used in categorical dummies and all categories are significant with a coefficient that increases linearly as the case size increases. The t-values are also all very high (t-values of three and above indicate statistical significance). The proportion of children in the case is significant with increasing coefficients across the categories, but it is a less important variable than case size. After controlling for case size, the removal of this variable leads to an R squared reduction of only 0.1 percent. Hence, case size predicts welfare much better than the share of children in the case.

The next set of variables relates to the socioeconomic characteristics of the PA. The occupation measured in terms of skill level is significant only for professionals, which increases welfare as expected. This variable contributes little to the overall model (0.1 percent), whereas the variable occupational category, which we also tested, turned out to be non-significant. This is understandable, because the occupation refers to the status of the PA before the crisis and not to the current occupational status. Hence the former occupation and, to a minor extent, the skills of the PA matter little for the current welfare of the case, indicating that returns to skills are very low for refugees. If refugees manage to work at all, the occupational employment is unlikely to match the skill level the refugees possess. This is evidently a non-negligible loss in human capital and productivity.7

The age of the PA is significant and contributes to the model by 0.2 percent. Older age increases welfare as compared to younger age, while there is little difference between the age classes of 35–54 and 55 and older. If the PA is of pension age, this is not a sign of reduced welfare for the case. Marital status is also a significant factor with a 0.2 percent importance, and if the PA is married or engaged then the case is more likely to be better off. Interestingly, being widowed reduces welfare as compared to being married but less so than being single, divorced, or separated. Education of the PA is not such an important variable in the economy of the model (0.1 percent) but it is clear that higher education is associated with higher welfare, with values that increase rather linearly as education increases. This is particularly true for university degrees and higher.

The following set of variables relates to migration patterns as initially explored in the poverty section. The place of origin is generally important (0.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.