The study of role and impact of formal education expenditures on economic growth in Iran

In this paper, we investigated the long run relationship between formal education expenditures and economic growth and also their influences on each other in Iran and for years 1979 to 2008. In this purpose vector autoregressive model (VAR) has been used. First, stable of variables by the use of Dickey-fuller test has been examined. Next, analysis of Johnson test for considering the convergence among variables has been used. The results of this research show that the variables employment divided by population, The initial level of per capita production of employers, employment, physical capital, human capital [Primary and secondary as a percentage of Gross domestic product (GDP)] and human capital (advance education as a percentage of GDP) have a positive effect on real product in Iran.


INTRODUCTION
Distance education has relation with national income.Human capital formation through investments of education leads to more skills of people and in this way increase saving growth and regarding that, productivity of capital goods have relation with verity of human capital and scientific and expert man, society needs trained high level manpower by economic progress in this way, education leads to increasing of product and national income and increased national income leads to more investment education and human capital.Formal education expenditures have been used in many articles.Blankenau and Simpson (2004) in their article titled "Public education expenditures and growth" explored this expenditure-growth relationship in the context of an endogenous growth model in which private and public investment are inputs to human capital accumulation.The positive direct effect of public education spending on growth could be diminished or even negated when other determinants of growth are negatively affected by general equilibrium adjustments.They showed that the response of growth to public education expenditures might be nonmonotonic over the relevant range.The relationship depended on the level of government spending, the tax structure and the parameters of production technologies.Jung and Thorbecke (2003) in their article titled "The impact of public education expenditure on human capital, growth, and poverty in Tanzania and Zambia: a general equilibrium approach" expressed that the impact of public education expenditure on human capital, the supply of different labor skills, and its macroeconomic and distributional consequences is appraised within a multisector CGE model.The model was applied to and calibrated for two heavily indebted poor countries (HIPCs); Tanzania and Zambia.The simulation results suggested that education expenditure could raise economic growth.However, to maximize benefits from education expenditure, a sufficiently high level of physical investment is needed, as are measures that improved the match between the pattern of educational output and the structure of effective demand for labor.An important *Corresponding author.E-mail: abouie.mehrizi@gmail.com.Tel: +98 9352995018.result of the simulation experiments is that a well-targeted pattern of education expenditure could be effective for poverty alleviation.Hwang (2005) in her article titled "asset distribution and tertiary education expenditure in developing countries" examined whether initial asset distribution affects the composition of government expenditure on education.Using a cross-section of developing countries, it is found that a country"s initial asset inequality is positively and significantly associated with the ratio of tertiary education expenditure to government education expenditure, even after controlling for some explanatory variables.Iacopetta (2011) in her article titled "Formal education and public knowledge" examined the transitional dynamics of an economy populated by individuals who split their time between acquiring a formal education, producing final goods, and innovating.The paper had two objectives: (i) uncovering the macroeconomic circumstances that favored the rise of formal education; (ii) to reconcile the remarkable growth of the education sector with the constancy of other key macroeconomic variables, such as the interest rate, the consumptionoutput ratio, and the growth rate of per capita income (Kaldor, 1961).
The transitional dynamics of human capital growth models, such as Lucas (1998), would attribute the arrival of education to the diminishing marginal productivity of physical capital.Conversely, the model proposed here suggested that it is the rate of learning that catches up with the rate of return on physical capital.As technical knowledge expands, the rate of return on education increases, inducing individuals to stay longer in school.The model's transitional paths are matched with long run U.S. educational and economic data.Gershberg and Schuermann (2001) in their article titled "The efficiency-equity trade-off of schooling outcomes: public education expenditures and welfare in Mexico" analyzed how a central government allocated resources to states in the education sector.In particular, they used two relevant criteria in the decision-making process: the equity-efficiency trade-off and unequal concern with respect to the characteristics of states.They performed empirical tests of Mexican state-level education expenditure by the Federal Government and examined changes in allocation patterns by comparing 1980 and 1990 cross sections.
A two-sector model was considered in a welfare maximizing context, which allowed for a theoretical as well as econometric solution for jointly determined educational expenditure and production.This joint modeling is important, and they provided for a straightforward and easily replicable solution.The addition of the roads sector provided an instrument for endogenously determined expenditure in schooling production.They found that the Federal Government trades some efficiency for gains in equity, but in doing so treats states differently, and that results had changed over time.Sylwester (2002) in her article titled "Can education expenditures reduce income inequality?"examined whether devoting more resources to education could positively affect the distribution of income [as measured by the Gini coefficient (Deininger and Squire, 1996)] within a country.From the findings, public education expenditures appeared to be associated with a subsequent decrease in the level of income inequality.This founding was robust to the inclusion of various control variables and appeared to be larger in high income nations.The founding suggested that devoting more resources to education might be one way to reduce the level of income inequality within a country.
The rest of the paper is organized as follows.Subsequently, the paper analyses previous studies.It then describes the data and the econometric methodology.Thereafter it discusses the results that emerge from the estimations.Finally, the conclusions of this paper are then presented.

DATA AND METHODOLOGY
We use this data from 1979 to 2006 of Iran.We found them in Central Bank of Iran.One vector autoregressive (VAR) model which possess k as exogenous variable, and p as time`s inhibition for each variable, in shape matrix is shown as following: In this relation,   and it`s lags, k×1 vectors are related to models variables.  , i= 1, 2,…, p are model`s coefficients for k×k matrix and   , k×1 vector is related to terms of model`s error.Now for linking short term behavior of   to long term balance values, we can bring above relation as vector error correction model as following: Where: . Matrix π contains of information of long term balance variables.
We follow the Johansen approach in determining long-run relationships.Patterson (2000) and Doornik and Hendry (2001) provide a full treatment of the issues involved in this method.The first step is to estimate the VAR in levels with an appropriate lag structure.The next stage involves determining the cointegrating rank, that is, the number of long-run equilibrium relationships or cointegration vectors among the variables.Finally, to allow a reasonable interpretation of the results, cointegration vectors are identified (Abouie and Safdari, 2011).

Theoretical principles
The model which is used for Evaluate the role and impact of formal education expenditures on economic growth in Iran is defined as following:  Where:: Real product, V: Employment divided by population,   : The initial level of per capita production of employers, : Employment, K: physical capital, H: human capital (Primary and secondary as a percentage of GDP) and HE: human capital (advance education as a percentage of GDP) Table 1.

FINDINGS AND DISCUSSION
We use the previous formulation to estimate a VAR model containing five variables.In order to fitness of VAR pattern first, it is necessary to investigate the persistency of variables.One of the common examinations which nowadays use for recognition of persistent of one time series process is unit root test; we can do this examination in two ways: Dickey Fuller"s and Test, Dickey Fuller"s generalized test.The results of the test for the variables in levels are presented in Table 2.
The results reported in Table 2 show that all of variables are I (0).After investigation of persistency of variables, one of the important stages in evaluation of vector regression model is choosing rank of pattern.For choosing optimum rank of pattern, we can use criterion of Akaike (1973) or Schwarz (1978).The most lag which is given to model is 2, and considering Table 3, the least quantity of Schwarz statistic is prepared in first lag, we can indicate that the optimum lag of VAR model is equal to 1.
In this article, we follows vectors and accumulated vector among variables of real product, employment divided by population, The initial level of per capita production of employers, employment, physical capital, human capital (Primary and secondary as a percentage of GDP) and human capital (advance education as a percentage of GDP) have a positive effect on real product in Iran.By the use of Johansen"s (1940) method, considering stationary test, variables which are under consideration are I (0).In Johnson"s method after doing necessary calculations for studding existence of convergence we use two criterions consist of  max and  trace .
If existence of convergence among variable is verified, we can say that balance and long term relation among variable is established.Results which are concluded from effect's examination and examination of maximum specific values for determination of accumulated vectors  4 and  5. Results of maximum of specific values for determination magnitude of accumulated vector are reported in Table 4.The magnitudes of vectors which are prepared by statistic of examination effect matrix are equal to 2 vector and magnitudes of vectors which are prepared statistic of maximum specific values are equal to 1. Considering that examination of maximum specific values is stronger than examination of effect matrix.Therefore, for determination of magnitude of accumulated vector, examination of maximum specific values is used.Considering results of Tables 4 and 5 in level of probability of 95% magnitude of long term relations among variables compatible pattern with economic theory is equal to (r=1)1 is determined.
In Table 6, number inside parentheses are statistic of accounting t. estimated coefficients of all variables in a meaning full level, 5% are significant from statistical aspect.Considering prepared results within investigated period, variables of the results of this research show that the variables employment divided by population, the initial level of per capita production of employers, employment, physical capital, human capital (Primary and secondary as a percentage of GDP) and human capital (advance education as a percentage of GDP) have a positive effect on real product in Iran.

Conclusion
Generally, in this paper we investigated the long run relationship between formal education expenditures and economic growth in Iran.In this article, first we presented a model and estimated this model, in order to fitness of VAR pattern we used unit root test, then the magnitude of inhibition of VAR model was determined after that by using of Johansson"s and existence of accumulated vectors showed long term relations among variables.After certainty about existence of long term relation, we estimated this relation and then interpreted these coefficients.
The results of this research show that the variables employment divided by population, the initial level of per capita production of employed person, occupation, physical capital, human capital (primary and secondary) and human capital (higher education) have a positive effect on real product in Iran.
Human capital has a positive relation with gross national product, the reason is human capital is a result for increasing country production power and physical capital also has a positive relation with real output.Because increased physical capital makes scientific manpower able to increase product by using technology and increase economic growth.
Employment rate has a positive relation with economic growth; it means that whatever employment level increase more manpower involved in product and increased manpower leads to increased efficiency of product and through this, economic growth increase.

Table 1 .
Variable definitions and descriptions.

Table 2 .
ADF tests for unit roots.

Table 3 .
Determination of magnitude of lag of VAR model.

Table 4 .
Test statistics for Co integrating rank (Trace tests).

Table 5 .
Test statistics for Co integrating rank (max tests).