Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/203035 
Year of Publication: 
2018
Series/Report no.: 
LIS Working Paper Series No. 749
Publisher: 
Luxembourg Income Study (LIS), Luxembourg
Abstract: 
As health care costs rise, so too does the importance of assessing their incidence, and factoring these costs into measures of post-government income distribution. This paper contributes to this assessment by calculating the effect of government policy on the distribution of income by adjusting income not only for taxes paid and social transfers received, but also for households' health expenditures. Standard measures of the effect of government policy on the distribution of income inconsistently accounts for these expenses; such inconsistencies result in biased measures of post-government income distribution. Here we account as much as possible for this inconsistency to provide more accurate cross-national comparisons of governments' redistributive effort. Using eight nationally-representative household datasets from 2010, we modify post-government income by treating households' health care costs similarly across countries. The results show the degree of bias in common estimates of the distribution of disposable income. In Switzerland and the US, for instance, estimates of post-government poverty rates climb by three to four percentage points once we account for households' medical expenses. We find that including private health care spending in measures of countries' redistributive effort results in greater variation among countries in their redistributive effort. We conclude that future assessments of governments' redistributive effort should uniformly account for the burden nations' health care financing policies place on households.
Subjects: 
Gini
health finance
inequality
Reynolds-Smolensky
international comparison
poverty
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.