for eg:
debt/ gsdp (gross state domestic product)
tax/ gsdp (gross state domestic product)
debt/ gsdp (gross state domestic product)
tax/ gsdp (gross state domestic product)
Fiscal space = [Total Revenue receipts – (Revenue expenditure on Interest Payments and Servicing of Debt + Revenue expenditure on Administrative Services) + Loans from the Centre] / Total Revenue receipts.
Intergovernmental transfers/ total revenue (total revenue is a combination of own revenue + intergovernmental transfers) is one of the methods of measuring level of dependency of sub national governments on higher level of governments for funds. In this case both vary? What is the remedy for this?
Is it Intergovernmental transfers/ total expenditure? Here also both Denominator and Numerator vary?
What is the justification given in the literature in which these type of variables are widely used in the analysis?
please reply or suggest material to refer.
Dear Dharshini,
There is one obvious justification for doing the ratios. If you are studying multiply regions, states, or countries, then the denominator will adjust for differences in size. If you are getting different results for the total revenue and expenditure in the intergovernmental transfers, then you should take the one with the best statistical fit, or report and discuss both if their significances are mixed. These are not so called "great ratios" in economics or finance, so do not expect to find literature that are definitive about them. Of course, you must have some apriori specification of how these variables are related in a model, but you statistical results will help you choose.
If you use a common denominator in regressions, consider potential problems due to spurious correlation between the ratios. This can be solved by spliting ratios in two variables via log transformations as in Kronmal, Richard A. "Spurious Correlation and the Fallacy of the Ratio Standard Revisited." Journal of the Royal Statistical Society. Series A (Statistics in Society) 156, no. 3 (1993): 379-92. doi:10.2307/298306. Or use inverse hyperbolic sine transformations if you have zeros in variables.
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication