Abstract:
The tax-expenditure hypothesis posed by Milton Friedman emphasizes a positive causal relationship between government tax revenues and government expenditures. If citizens do not have a correct perception of the real tax burden and under-estimate the price of public goods and services, there is a negative causal relationship between tax revenues and government expenditures, which indicates existence of fiscal illusion. Using quarterly data. for the period 2001-2012, this paper investigates Fiscal illusion in Iran. In order to achieve this goal, two symmetric and asymmetric error correction models, are estimated. According to results from Wald test in symmetricmodel,there is a negative causal relationship between real tax revenues, and real government expenditures. This result hence, confirms fiscal illusion in Iranian economy.Also, the results obtained from the asymmetric model show that there is merely fiscal illusion in the case of tax revenues reduction and there is no Granger causal relationship for the positive changes of tax revenues. Therefore, by a decline in tax revenues, government expenditures increase after a year due to fiscal illusion. Thus it seems that in the state of government's budget deficit, raising the taxesis an efficient instrument.
Machine summary:
Studies carried out by Wagner (1976), Buchanan and Wagner (1977) and Niskanen (1978, 2002, 2006) demonstrate that fiscal illusion leads to a negative causal relationship between tax revenues and government expenditures, though this issue contradicts the traditional hypothesis posed by Friedman (1978).
2. Seasonal Unit Root Test In order to test fiscal illusion in Iranian economy, based on government expenditures and tax revenues data, first the stationary of time series data should be investigated.
Vector Error Correction Model As the two variables, government expenditures and tax revenues, are cointegrated, in order to investigate the causal relationship between them, an error correction model is used, following the carried out studies by Bohn (1991), Mounts and Sowell (1997), Koren and Stiassny (1998), Garcia and Henin (1999), Ewing at al.
In this research, as the variables are seasonal; the seasonal vector error correction model (SECM) is applied and according to the aim of this survey which is testing the traditional hypothesis of positive causal relationship between tax revenues and government expenditures versus the hypothesis of fiscal illusion, estimating the below regression is concerned: p p 2Gt i 2Gt i i 2Tt i 1wˆ t 1 2vˆt 1 ut (7) i1 In this equation, i1 vˆt 1 and wˆ t 1 are respectively the residuals of cointegrated equations (3) and (4).
In this research, using seasonal data of government expenditures and tax revenues from during 2001-2012 fiscal illusion hypothesis in Iranian economy was tested.