Asian Research Consortium

Transcription

1 Asian Research Consortium Asian Journal of Research in Business Economics and Management Vol. 5, No. 9, September 2015, pp ISSN Asian Journal of Research in Business Economics and Management Cointegration and Causality Analysis of Public Expenditure and Growth with Special Reference to Rajasthan Kirandeep Kaur*; Dr. Hemlata Manglani** Abstract *Research Scholar, Department of Economics, Central University of Rajasthan, Ajmer, Rajasthan, India. **Assistant Professor and Coordinator, Department of Economics, Central University of Rajasthan, Ajmer, Rajasthan, India. DOI NUMBER / X The present study empirically analyzed the relationship between State Government Expenditure and Net State Domestic Product for Rajasthan Economy from period 1980 to The study utilizes various econometrics techniques of the time series data analysis such as ADF Test for Unit Root, Engle and Granger method and Johansen techniques for Cointegration and to find the causality between both the variables the study employed the Vector Error Correction Method. The results of the study revealed that there is long run relationship between both the variables. The findings of the Vector Error Correction Method states that there is unidirectional causality between both the variables and the direction of the causality is from Net State Domestic Product to State Government Expenditure which implies that the Rajasthan Economy follow the Wagner law of the Public Expenditure reveals the Economic Growth is the causal factor to the Growth of the Government Expenditure. Keywords: Wagner s Law, Public Expenditure, Unit root, Growth, Net State Domestic Product, Cointegration, Causality analysis. 26

2 1. Introduction The relationship between government expenditure and economic growth has been widely discussed by researcher in the field of public finance. Government plays a significant role in any economy. According to Keynes the government is the only balancing factor in the economy which can balance the economic system with the use of fiscal policies. After the great depression of 1936, Keynes was the economist which drew the attention for the importance of the government in the economic system of the country. According to the General theory of Keynes, in case of depression the government should come up with the expansationary fiscal policy to increase the money supply as well as to increase the demand for goods and services which will leads to the sustain growth in the economy. According to Keynesian view of Public expenditure to enhance the economic growth the government must increase the level of public expenditure. Public expenditure is a tool without which the engine of the economic growth could not work. In short the Keynesian view of public expenditure considers that the public expenditure is the exogenous factor and has positive and significant impact on economic growth of the country. In contrast to the Keynesian view of the public expenditure the Wagner was in favour to increase the economic growth and states that it s the economic growth which leads an increase in the level of public expenditure. In other words according to the Wagnerian law of public expenditure as the economic growth or the real per capita income of the country increases the share of public expenditure in total national income will also increases. The Wagner law of increasing state activities revealed that the economic growth is the exogenous factor and it is the cause to increase in the share of public expenditure in total national product of the country. There has been huge literature on the relationship between the public expenditure and economic growth. The different study has the different results with the use of different time period. Some study found the relationship in favour of Keynesian hypothesis (Govindraju et al. 2010), some are in favour of Wagner Hypothesis of Public expenditure (Abdullah, Maamar, 2010, Bagdigen, Centintas, 2013, Ankitoboy et al. 2006, Demirbas, 1999, Oxley, 1994), some study concluded that there is both way relationship and also some states that there is not any relationship between these two variables (Afzal, Abbas 2010, Ansari et al 1997, Burney 2002, Chimobi 2009). The aim of the present study is to analyze the relationship between public expenditure and Net State Domestic Product in case of Rajasthan in both total and per capita term. The rest of the paper is structured as follows. Section 2 reviews literature on relationship between government spending and economic growth. Section 3 presents picture of Rajasthan economy. Section 4 provides model and methodological framework of the study. The Section 5 provides the empirical findings of the study and The Section 6 summarizes the major findings of the study and conclusion of the study. 2. Review of Literature Schallegger and Torgler (2004) examined the effects of government expenditure on state and local level. The study used the data from for panel analysis. The results of the panel analysis inferred the negative relation between government size and economic growth. Fan et al. (2004) explored the effects of different types of government expenditure on agriculture growth and rural poverty Uganda by utilizing the district-level data. The study concluded that expenditure on agricultural research and extension and rural roads has positive impact on agriculture production and rural poverty reduction respectively. Kuhar et al (2005) analysed the impact of public 27

3 expenditure on economic growth in the Solvenia with the use of the input-output analysis. The finding shows that public expenditure has the positive impact on economic growth in Solvenia. John and George (2005) determined the causality between the share of public expenditure in GNP and economic growth. The study used the annual time series data for Greece, UK and Ireland. The empirical findings of the study indicate that Government expenditure Granger causes economic growth in UK and Ireland in the short run as well as in the long run while the causality was in inverse direction in case of Greece. Ahmad and Ahmad (2005) investigated the long run association between government size and per capita income in D-8 member countries. The study utilized the various time series econometrics techniques. The result of the empirical analysis of the study reveals that there is no relationship between government expenditure and per capita income in all the countries. The results of the causality analysis states that there is short run bi-causality between both the variables in case of Iran. Huang (2006) conducted a study to test the validity of Wagner law for China and Taiwan. The author employed the bound test and Toda-Yamamoto s Granger Non Causality test to analyze the time series data from 1979 to The empirical findings failed to find the long run relationship between both the variables. The study concluded that the Wagner law does not exist in both the countries. Tilak (2006) analyzed the trend in education expenditure of Rajasthan and Andhra-Pradesh.The study concluded that the performance of Rajasthan was much better than the Andhra Pradesh and if this situation will remain continuous then the Andhra Pradesh may replace Rajasthan in the category of BIMARU State. Karpetis (2006) made an attempt to develop a simple dynamic Keynesian model to find out the impact of changes in the level of government expenditure and growth rate and money supply on various macroeconomic variables. The results confirmed that the long run values of inflation are affected by the size of government spending and money supply. Olugbenga and Owoye (2007) conducted cointegration and causality techniques to analyze the dynamic link nexus between government expenditure and economic growth for a group of 30 OECD countries with the annual data from period 1970 to The empirical results of the study revealed a unidirectional causality from government expenditure to growth for only 16 countries which implied that these economies have the evidence in supports of the Keynesian hypothesis. While out of 30 countries the 10 countries were found with the causality from economic growth to government expenditure i.e. the existence of Wagner s law. The study observed that remaining four countries have the evidence of a bi-causal relationship between government expenditure and economic growth. Bose et al. (2007) found that the share of government capital expenditure in GDP is positively and significantly associated with growth whereas the current expenditure is insignificant. The government expenditure on education and total expenditure on education are significantly associated with economic growth. Sideris (2007) examined the hypothesis of Wagner law of increasing states activities for Greece. The result of the empirical analysis states the existence of long run relationship between government expenditure and economic growth. The findings of the causality analysis confirmed that causality runs from national income to government expenditure. Lamartina and Zaghini (2011) found the evidence of long run positive relation between government expenditure and economic growth in 23 OECD countries. Omoke (2009) analysed the direction of causality between Government expenditure (GE) and National Income (NI) in Nigeria with the use of annual data of 35 years from 1970 to The empirical analysis of the study exhibited that there was no long-run relationship between government expenditure and national income in Nigeria. The findings of the Granger causality test found the evidence of causality from government expenditure to national income so the study 28

4 recommended that government expenditure plays a significant role to promote the economic growth in Nigeria. 3. Trends of Public Expenditure and Growth in Rajasthan Rajasthan is an agriculture based economy and mainly facing the problem of Famine and drought. The Rajasthan states have the largest area in all the states of the Indian but it has the lower rank in case of economic development and the per capita income in all the states of the India. the Compound annual Growth rate of gross state domestic product of Rajasthan state is recorded 19 percent form to The Net state Domestic product and the government expenditure of the state are continuously increasing in both total as well as per capita term. The most of the part of the state is facing the problem of drought in continuous years and for which the state government has to spend a larger part of the budget on relief program in those area. So a large share of the public expenditure goes to these relief programs which are not considered as a development expenditure of the states. The table 1 gives the Total government expenditure and Net State Domestic Product of the state which shows that the both the government spending as well as the net state domestic product of the Rajasthan state has the increasing trend Over the years. Table- 1. Government Expenditure and Net State Domestic Product of Rajasthan (in Lakh Rupees) Year TGE NSDP % Share of Public Expenditure in NSDP Source: Author s Computation with the use of various issues of Economic Review of Rajasthan and Statistical Abstract of Rajasthan. The figure 1 shows the net state domestic product and government expenditure of Rajasthan state. The graph shows that there is increasing trend in the net state domestic product and government expenditure of the state from 1980 to

5 4. Data and Methodology From the review of Literature the present study estimated the following two model to find the relationship between government expenditure and economic growth in both total and per capita term- ln(ge) = f ln(ec) ln(ge/ Population) = ln(eg/ Population) Here GE is stand for total government expenditure and EC for economic growth. For the empirical analysis of the objectives the Present study utilizes the annual time series data of State Government Expenditure and Net State Domestic Product from 1980 to All the data is taken from statistical abstract of Rajasthan. Raw data converted into logarithmic form to get the stationary at lower level and also to remove the problem of Heteroscadasticity and Autocorrelation of the data. To remove the impact of price change the data has been converted into real term with the use of Net State Domestic Product Deflator. Before the analysis the study used the ADF unit root test to test the inherent time series properties of the data with constant and with constant and intercept. To investigate the long run relationship between the variables the study used the engle granger two step methodology according to which if the residuls series of the model is integrated of order zero then is said to be cointegrated. If all the variables of the study are integrated of same order then the johansen Julisus test of cointegration can be used to test the cointegration between the variables. The study also utilized the Johansen and Julisus test for cointegration test.after the cointegration analysis the study employed the Vector Error Correction Method to study the direction of causality between the variable and the following equation has been estimated for the VECM Model- 30

6 ln GE ln GE ln NSDP ECT 1t 10 11, i 1, t1 12, i 2, t1 1 t1 1t i1 i1 ln NSDP ln GE ln NSDP ECT 2t 20 21, i 1, t1 22, i 2, t1 2 t1 2t i1 i1 ln PCGE ln PCGE ln PCNSDP ECT 1t 10 11, i 1, t1 12, i 2, t1 1 t1 1t i1 i1 (1). (2) (3) ln PCNSDP ln PCGE ln PCNSDP ECT 2t 20 21, i 1, t1 22, i 2, t1 2 t1 2t i1 i1. (4) The VECM model estimates both the short run as well as long run causality between the variables. To test the efficiency of the VECM model the study used the Normality Test, Serial Correlation LM Test and ARCH Test for heteroscadasticity. The study also utilizes the CUSUM stability test for checking the stability of the model. 5. Empirical Results and Discussion Table-2 represents the results of the unit root ADF Test for the stationary prosperities of the data. The results of the ADF Test states that all variables are non-stationary al level and stationary at first difference. Table-1 Unit Root Test Result Variable Level First Difference With intercept With intercept and trend With intercept With intercept and trend lnge * * lnnsdp * * lnpcge * * lnpcnsdp * * (*) indicate significant at 5 percent level. 31

7 Table -3. Results of Engle Granger Cointegration Method First Step- Model-1 Variable Coefficient Standard Error t-value C * lnnsdp * R-Square DW Stat Adj-R Square F-stat (0.000) Model- 2 Variable Coefficient Standard Error t-value C * lnpcnsdp * R-Square DW Stat Adj-R Square F-stat (0.000) Second Step Model Variable With intercept With intercept and Trend 1 μ * * 2 μ * * (*)indicate significant at 5 percent level. The results of the Engle Granger Cointegration method reported in table 3 states that there is cointegration between the variable or the variables have the long run relationship in both the models of the study. Table-4. Results of Johensen Cointegration Test Test H0 Eigen Value Statistics 5 % CV P-Value Model - 1 Trace r= * r= Max. r= * r= Model- 2 Trace r= * r= Max. r= * r= (*)indicate significant at 5 percent level. 32

8 Table 5. Results of VECM Long Run Causality Test Direction of Causality ECT Coefficient Standard Error t-value P-Value LnNSDP lnge ECT * LnGE lnnsdp ECT LnPCNSDP ECT * lnpcge LnPCGE lnpcnsdp ECT (*)indicate significant at 5 percent level. The results of the VECM Model presented in table 5 states that there is long run causality between net state domestic product and government expenditure and the direction of the causality lie from net state domestic product to government expenditure and the inverse of the causality is not significant which implies that there is unidirectional causality from net state domestic product to government expenditure. The results of the second model states that there is one way causality between per capita government expenditure and per capita net state domestic product and the direction of the causality is from per capita net state domestic product to per capita government expenditure. Table- 6.Results of Short Run Causality from VECM Model Model Direction of the Causality F-Stat. Chi-Square Value Model-1 LnNSDP lnge * LnGE lnnsdp Model-2 LnPCNSDP lnpcge * LnPCGE lnpcnsdp (*)indicate significant at 5 percent level. The results of the short run causality from VECM Model or the results of the Wald Test are reported in table. The results reveals that the there is absence of the causality between the variables in both the model. It states that in short run there is no causal relationship between both the variable in gross and per capita term. Table- 7. Results of Diagnostic Testing Direction of Causality Normality Test ARCH Test Serial Correlation LM Test JB Stat. Obs* R 2 Obs* R 2 Model-1 LnNSDP lnge LnGE lnnsdp Model-2 LnPCNSDP lnpcge LnPCGE lnpcnsdp

9 Table- 8Summary Statistic of the Variables LNGE LNNSDP LNPCGE LNPCNSDP Mean Median Maximum Minimum Std. Dev Skewness Kurtosis Observations The results of Diagnostic test of both the model are presented in table-7. The results of the diagnostic test states that there is no problem of Normality, Heteroscadasticiy and Serial Correlation in both the model which states that the model is good fit to data and the model is efficient for the analysis. The results of the summary statistics are reported in table-8. The results of the summary statistics states that the value of the Skewness is near to zero and the Value of Kurtosis is near to the 3 which is the indicates of the normality distributed data. 15 Figure -2. Results of Stability Test Model-1 CUSUM Test Result CUSUM 5% Significance Model-2 CUSUM Test Results CUSUM 5% Significance 34

10 The results of the stability CUSUM Test shows that the estimated model does not have any instability. The model is good fit in case of the stability criteria of the model. 6. Conclusion The study provides the empirical evidence of the relationship between State Government Expenditure and Net state domestic product of the Rajasthan economy with the use of 34 years annual time series data from 1980 to To analyze the objective of the study the various econometrics time series techniques has been utilized like ADF Test, Engle Granger Two Step Cointegration Method, and Johansen Cointegration Technique. To find the causality between both the variables the study employed the Vector Error Correction Model. The result of the study indicates that there is long run relationship between both the variables. The empirical findings of the study states that there is one- way causality between both the variables and the direction of the causality is from Net state domestic product to state government expenditure. The findings of the study have the evidence in favour of the Wagnerian hypothesis of public expenditure (causality from economic growth to government expendsiture). The study concluded that the Rajasthan economy follow the Wagnerian View of the state expenditure. The findings of the study are somewhat similar to the findings of the Akpan (2011), Omoke (2009), Reddy (1972) and Singh & Sahini (2002). References Abdullah, H., & Maamor, S. (2010). Relationship between National Product and Malaysian Government Development Expenditure: Wagner s Law Validity Application: International Journal of Business and Management, 5(1): Afzal, M., & Abbaa,Q. (2010). Wagner s Law in Pakistan: Another Look. Journal of Economis and International Finance, 2(1): Ahmad, Naved & Fareed Ahmad (2005). Does Government Size Matter? A Case Study of D-8 Member Countries. Pakistan Economic and Social Review, 43(2): Akitoby B., Clements B., Gupta S. & Inchauste G. (2006). Public Spending, Voracity and Wagner s law in developing countries. Eur J Polit Econ., 22(4): Akpan, U.F. & Abang D.E. (2013). Does Government Spending Spur Economic Growth? Evidence from Nigeria. Journal of Economics and Sustainable Development, 4(9): Ansari M.I., Gordon D.V., Akuamoah C. (1997). Keynes versus Wagner: Public Expenditure and National Income for three African countries. Appl. Econ., 29(4): Bağdigen, Muhlis & Hakan Çetintaş (2003). Causality between Public Expenditure and Economic Growth: The Turkish Case. Journal of Economic and Social Research, 6 (1): Bose Niloy, M. Emranul Haque & Denise R.Osborn (2007). Public Expenditure and Economic Growth A Disaggregate Analysis for Developing Countries. The Manchester School, 75 (5):

11 Brown, R. L., Durbin, J., & Evans, J. M. (1975). Techniques for Testing the Constancy of Regression Relationships Over Time. Journal of Royal Statistical Society, 37: Burney, N.A. (2002). Wagner s Hypothesis: Evidence From Kuwait Using Cointegration Tests. Applied Economics, 34(34): Demirbas, S. (1999). Cointegration Analysis-Causality Testing and Wagner s Law: The Case of Turkey typescript, May. Fan, S., X. Zhang & N. Rao (2004). Public Expenditure, Growth and Poverty Reduction in Rural Uganda. Development Strategy and Governance Division Discussion Paper No.4.Washington, DC. International Food Policy Research Institute. Govindaraju, C.V.G.R.,Rao,R., & Anwar, S.(2010). Economic Growth and Government Spending in Malaysia: A Re-examination of Wagner and Keynesian View. Economic Change and Restructuring. Gujarati D.M. (2003). Basic Econometrics 4th edn. McGraw-Hill/Irwin. New York. Huang, Chiuung-Ju (2006). Government Expenditure in China and Taiwan: Do They Follow Wagner s Law. Journal of Economic Development, 31(2): John, L. & George, V. (2005). Government Expenditure and Economic Growth: Evidence from Trivariate Causality Testing. Journal of Applied Economics, Vol. VIII, No. 1: Karpetis, C., (2006). Inflation Targeting and the Multiplier Accelerator Principle. Journal of Applied Business Research, 22(4) : Kuhar, A., Juvancic. L.,Sila. U., & Erjavec. E. (2005). Evaluation of Public Expenditure on Economic Growth of The Peripheral Slovenia with Input-Output Model. Acta Agriculture Slovenia, 86 (1): Lamartina, S. & Zaghini A. (2011). Increasing Public Expenditures: Wagner s Law in 7 OECD Countries. German Economic Review,12: Olugbenga, A.O. & Owoye, O. (2007). Public expenditure and economic growth: New evidence from OECD countries. Business and Economic Journal, 4(17). Omoke, P. (2009). Government Expenditure and National Income: A Causality Test for Nigeria. European Journal of Economic and Political Studies, 2: Oxley, L. (1994). Cointegration, Causality and Wagner's Law: A Test for Britain Scottish Journal of Political Economy, 41(3): Mann, A.J. (1980). Wagner s Law: An Econometric Test for Mexico National Tax Journal, 33(2):

12 Reddy, K.N. (1988). Determinants of Growth of Government Expenditure in India. Indian Journal of Public Administration, 34(1) : Schaltegger, C.A. & B.Torgler (2004). Growth Effects of Public Expenditure on State and Local Level: Evidence From a Sample of Rich Government. Centre for Research in Economic. Managerment and The Arts. Sideris, D. (2007). Wagner s Law in 19 th Century Greece: A cointegration and Causality Analysis. Bank of Greece Working Papers.64. Statistical Abstract of Rajasthan. Directorate of Economic and Statistics, Rajasthan Jaipur to Tilak, J.B.G. (2006). Trends in Public Expenditure on Education: A Contrast between Two Educationally Backward States- Andhra Pradesh and Rajasthan. Journal of Indian School of Political Economy,18 (1-2):1-34. Singh, B. & Sahni, B.S. (1984). Causality between Public Expenditures and National Income, Review of Economic and Statistics. 66(4):