Investigating the Effect of Government Financing Methods on Economic Growth in Iran: Markov-Switching (MS) Approach

Introduction: It is important to note that the effects of different methods of government financing (such as taxes, borrowing, selling natural resources, etc.) are not the same in the economy, and these different methods can affect macroeconomic variables such as economic growth in different ways. Therefore, answering the question of what is the optimal government-financing portfolio in the economy is of particular importance, especially in developing countries where the sale of natural resources plays an important role in their financing portfolio. According to Sachs and Warner, the abundance of natural resource encourages rent seeking, corruption, and poor government management. It also encourages developing countries to engage in protectionist paths through state-led projects of development, in fear of “Dutch disease effects of the resource abundance”. Yet, as de Ferranti et al. (2002) put it, ‘it is impossible to argue that Australia, Canada, Finland, Sweden and the United States did not base their development on their natural resources.’ (p. 6)Accordingly, considering the three main current issues of Iran's economy based on a) inefficiency of the tax system, b) dependence of the budget on oil, and c) growing government debts, this paper investigates the effect of government financing methods on the economic growth in Iran from 1973 to 2018 using a Markov-Switching (MS) model.Methodology: Following the generalized growth accounting model based on neoclassical and endogenous growth models, to investigate the effect of different methods of government financing on the economic growth, we used the following model based on the concept of the total production function: where:GY: GDP growth rate (constant 2011 LCU),GK: Gross capital formation growth rate (constant 2011 LCU) as a proxy for investment growth rate,GL: Population growth rate as a proxy for labor growth rate,GX: Export growth rate (constant 2011 LCU) as a proxy for export growth rate,TR / Y: Government tax revenue as a percentage of GDP,OR / Y: Government oil revenue as a percentage of GDP,GD / Y: Government debt as a percentage of GDP.In addition, this study uses annual time series for Iran during 1973-2018.Results and Discussion: Based on the specification tests, we estimated MSI (2) model using the EM algorithm as reported in Table 1. This model was tested for linearity using the LR linearity statistics assuming the null and alternative hypotheses to be a linear model and an MS model, respectively. The probability value of the Chi^2 statistic in this test (0.005) supports the existence of non-linearity in the data. Based on the transition probabilities, the probability of moving from regime zero (one) to one (zero) regime is 0.6879 (0.6374). Therefore, it can be said that the probability of staying in both regimes is moderate. Table 1. Results from MSI (2) Coefficient t-value t-probSwitching variables in Regime 0Intercept0.600.2940.771GY (-1)-0.04-0.810.423TRY0.110.380.704ORY-0.18-3.250.003GDY-0.44-5.280.000GDY (-1)0.263.190.004Switching variables in Regime 1Intercept1.470.690.496GY (-1)-0.30-5.020.000TRY0.86 2.540.017ORY-0.10-1.840.076GDY-0.70-8.800.000GDY (-1) 0.51 7.550.000Non- switching variablesGK0.137.310.000GL4.094.860.000GL (-1)-1.24-1.300.204GX0.2416.10.000AIC = 5.33 SC = 6.09Linearity LR-test Chi^2(9) = 42.361 [0.000]p_ {0|0} = 0.6879; p_ {1|1} = 0.6374p_ {0|1} = 0.3121; p_ {1|0} = 0.3626 Normality Test: Chi^2(2) = 0.08 [0.959] ARCH 1-1 Test: F (1,26) = 1.13 [0.298] Portmanteau (6): Chi^2(6) = 10.46 [0.106] Note: In this study, in order to determine the optimal lag of variables, the autoregressive distributed lag (ARDL) technique was used. Also, the unit root tests results showed that all the variables were stationary.* Annual data for all the variables were obtained from the Central Bank of Iran.Source: Research findings Figure 1. Regime classification based on the filtered and smoothed probabilitiesSource: Research findings In addition, the cumulative effects of explanatory variables are presented in Table 2: Table 2. Cumulative effects of explanatory variables on economic growth in zero and one regimescoefficient in regime onecoefficient in regime zerovariable0.1250.1GK2.742.19GL0.230.18GX0.110.66TR/Y-0.17-0.08OR/Y-0.17-0.14GD/YSource: Research findings Conclusion: This paper investigates the effect of government financing methods on the economic growth in Iran from 1973 to 2018. To this end, a Markov-Switching (MS) model was used. The results showed that the tax-to-GDP ratio had a positive effect on the economic growth in both recession and boom regimes, although this effect was not significant in the recession regime. Given the low tax-to-GDP ratio in Iran, these results have not been unexpected. Also, based on the findings of this study, the ratio of government oil revenues to GDP had a negative and significant effect on economic growth in both identified regimes, which confirms the theory of resource curse phenomena or paradox of plenty in the Iranian economy. In addition, the findings of this study showed that the ratio of government debts to GDP had a negative and significant effect on the economic growth in both regimes. The reason for this negative impact could be the fact that government borrowing in Iran is used to compensate for structural budget deficits instead of spending on productive investments and building the necessary infrastructure. Finally, the findings showed that the investment growth rate, population growth rate and export growth rate had positive, significant and tangible effects on Iran's economic growth, respectively.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.