Before implementing the graphical causal technique and FEVD, the important specification issue to be addressed is the determination of a correctly specified vector autoregressive model (VAR) in equation. In the general VAR system with n variables, if all the variables are stationary, the appropriate modeling strategy is to estimate a (unrestricted) VAR in levels (i.e., equation ). If all the variables are all nonstationary / and there is no cointegration at all, then the appropriate model is a VAR in first differences involving no long-run elements. If all the variables are cointegrated, on the other hand, then one can model the system as a vector error-correction model (VECM) (Harris and Sollis 2003). Since time series data are most likely nonstationary /(l) processes, the first step in determining a correctly specified VAR model is to identify such time series properties of the variables (i.e., nonstationarity and cointegration). The presence of a unit root in Zt in equation is first tested using the Dickey-Fuller generalized least squares (DF-GLS). The results show that the null hypothesis of nonstationary cannot be rejected at the 5 percent level for all the level series, but can be rejected for all the first difference series, indicating that all the series are nonstationary and integrated of order one processes (Table 1).

Continue reading

# Monthly Archives: November 2013

# A New Look at the Emergence of an Agro-Energy Nexus: A Graphical Causal Approach – Data

The other objective is to measure the relative contribution of a structural shock in each variable to fluctuations of its own and other variables. This VMA form represents Zt as a linear combination of current and past forecast errors. Equation can be viewed as a flexible dynamic approximation to the true but unknown underlying relationships in the structural form equation of A0 CZt = D Al CZt_l + ut, in which empirical regularities of all the dynamic interrelationships can be statistically captured in the reduced form equation of Zt = A0—1D Al CZt_l + CA— Cut. In this respect, the second procedure is to impose the identified contemporaneous causal structures on A0 in et = CA0—1 \2ut to orthogonalize the residuals (et) into the underlying structural shocks (ut) such that Cov (et) = A0—1 c(A0 J. Finally, each (i, j) th component of Cs (i, j) = F s C^0—1 represents the (impulse) response of the i th variable to a structural shock in the j th variable in s period. On the other hand, the (i, j)th component of [Cs (i, j)]2 measures the relative contribution of each structural shock in the j th variable to the forecast error variance of ith variable in speriod. Weekly data are collected for the period from the first week of July 2006 to the first week of July 2008. The data span is chosen because a plethora of studies indicates a consensus that crop prices surge mainly occurred during the period.The U.S. prices of corn, soybeans, wheat and ethanol, and are collected from the Chicago Board of Trade. The exchange rate is the weighted average of the foreign exchange value of the U.S. dollar against the major currencies and is taken from the Federal Reserve Board. The oil price is obtained from the Department of Energy. All variables are converted to natural logarithms and used throughout.

Continue reading

# A New Look at the Emergence of an Agro-Energy Nexus: A Graphical Causal Approach – The Model

**The Links between Crop prices, Exchange rates, Oil Prices, Biofuel Demand**

Since the focus of this study is the explanation of changes in prices for major U.S. crops, following previous studies, three factors which are found to be of central importance to influence those prices are selected for the model – that is, exchange rates, crude oil prices and biofuel demand. The reduced-form equations for crop supply and demand are specified as follows:

Of = Q- (p^ 5Ps5pw5Po 5 xf ), Vi = cornr soyheans and wheat

QD = QD (Pc, Pe, er, xD)

Of = Of (p,,ER,Xf ), Vi = soybeans and wheat

The supply equations for each crop (Qf) are specified as a function of prices of corn (Pc), soybeans (i*), and wheat (Fw), oil prices (P€.) and crop-specific supply shifters (A’f). Three prices in the equations represent the linkages among crops, while oil prices capture the cost-push effect of energy price changes on agricultural production costs. The demand equation for com (Q° ) is specified as a function of corn prices (Pc), ethanol prices (P0), exchange rates (ER) and crop-specific demand shifters (X^). In equation ethanol prices capture the demand-pull effect of corn-based ethanol production on corn prices. The demand equations for soybeans and wheat are specified as a function of its price, exchange rates and crop-specific shifters. Exchange rates in equations and capture the effects of changes in the foreign demand on crop prices.

The market equilibrium conditions for each crop are then:

Pi = Pi (P-i, Pe, P0, ER, Xf, XD)

where /’ ( = {/’, I\, I\ } \ {/’}, Vi = corn, soybeans and wheat. Equation postulates simultaneous systems of equations (SSE)to represent interrelationships among prices of the three crops, exchange rates, and prices of oil and ethanol. Note that ethanol prices are incorporated in the soybean and wheat price equations through its effect on the corn prices.

** Continue reading **

# A New Look at the Emergence of an Agro-Energy Nexus: A Graphical Causal Approach – Introduction

In the literature of agricultural economics, the sharp hikes in prices for major U.S. crops -that is, corn, soybeans and wheat- during 2006-2008 have received considerable attention in recent years/ln addition to numerous studies, briefings, and other materials financed or produced by national and international research institutes and interest groups, many scholars have independently studied this issue.The results from these studies indicate that, among other things, rising energy costs (i.e., crude oil prices), increased demand for biofuel production and the weak U.S. dollar have been the main culprits behind the rapid surge in major crop prices during 2006-2008. Trostle, for example, shows that the downward spiral of the value of the U.S. dollar has helped U.S. major crops more competitive in the world market and exerted upward pressure on prices of those crops through the enhanced foreign demand. He also adds that the phenomenal surge in crude oil price during 2007-08 has resulted in hikes in production costs and thus crop prices. Mitchell and Park and Fortenbery, on the other hand, find that the rise in crude oil prices has encouraged the rapid expansion of biofuel production in the U.S. under the Energy Security and Independence Act of 2007, thereby pushing up demand for farm commodities (i.e., corn) and prices.

Continue reading

# The Distributive Effects of Joining the Global Economy in Iran: the Application of ARDL Model – Conclusion

The short-run adjustment process is examined from the ECM. If the coefficient of ECM lies between 0 and -1, the correction to GINI in period t is a fraction of the error in period t-1. In this case, the ECM causes the GINI to converge monotonically to its long-run equilibrium path in response to the changes in the exogenous variables. If the ECM is positive or less than -2, this will cause the GINI to diverge. If the value is between -1 and -2, the ECM will produce dampened oscillations in the GINI around its equilibrium path.

Following Kremers, Ericson and Dolado (1992) who argued that the significant lagged error-correction term is a more efficient way of establishing co-integration, we concluded the existence of a strong co-integration relationship among variables in the model. However the coefficient of ECM(-1) is -0.90854, indicating that any deviation from the long-run equilibrium between variables is corrected about 90% for each period and suggests a high speed of convergence to equilibrium. Approximately 90% of disequilibrium from the previous year’s shock converge back to the long-run equilibrium in the current year.

Continue reading

# The Distributive Effects of Joining the Global Economy in Iran: the Application of ARDL Model – Empirical Results

Table 1 reports the results of unit root test applied to determine the order of integration among time series data. ADF Test has been used at level and first difference under assumption of constant and trend. Results clearly indicate that the index series except for trade intensity index are not stationary at level but the first differences of the logarithmic transformations of the series are stationary. Therefore, it can be safely said that series are integrated of order one I(1).

The next step is where equation 1 is estimated to examine the long-run relationships among the variables. As suggested by Pesaran and Shin (1999) and Narayan (2004), since the observations are annual, we choose 2 as the maximum order of lags in the ARDL and estimate for the period of 1977-2007. In fact, we also used the Schwarz-Bayesian criteria (SBC) to determine the optimal number of lags to be included in the conditional ECM(error correction model), while ensuring there was no evidence of serial correlation, as emphasized by Pesaran, Shin and Smith (2001). Table 2 indicates that macroeconomic variables significantly explain the Gini coefficient. The value of R-Bar-Squared is 0.96 which indicates a high degree of correlation among variables.

Continue reading

# The Distributive Effects of Joining the Global Economy in Iran: the Application of ARDL Model – Data and Econometric Methodology

There are several approaches to co-integration: e.g., the residual based Engle-Granger test, maximum likelihood based Johansen, and Johansen and Juselius test. These approaches require that all variable be integrated of the same order; otherwise create inefficiency which affects the predictive powers. Pesaran, Shin and Smith, developed the Autoregressive Distributive Lag Model or ARDL bounds testing approach to co-integration which is better suited to small samples.

Continue reading

# The Distributive Effects of Joining the Global Economy in Iran: the Application of ARDL Model – Literature

Wallace, Gauchat and Fullerton, examined the impact of five measures of globalization (global capital, foreign direct investment, exports, foreign born non-citizens, and foreign born citizens) and six measures of labor market transformation (deindustrialization, corporate restructuring, bureaucratic burden, casualization, bad jobs, and multiple job holding) on metropolitan-level earnings inequality of full-time, full-year workers 16 years and older. Their study makes several major contributions to the literature. First, they updated and extended the long line of studies on metropolitan earnings inequality. Second, they showed that these various dimensions of globalization and labor market transformation exert independent and mainly polarizing effects on the earnings distributions of metropolitan areas, net of controls for labor market structure and socio demographic variables. Third, they demonstrated the benefits of looking at the causes of inequality in the upper and lower tails of the earnings distribution. Finally, they developed a procedure to estimate counterfactual values of earnings inequality for all major metropolitan areas in the US in 2000. In the process, they provided a comprehensive accounting of the impact of globalization and labor market transformation on metropolitan earnings inequality.

Continue reading

# The Distributive Effects of Joining the Global Economy in Iran: the Application of ARDL Model – Review

Globalization and inequality is a highly debated topic in the literature. Various studies prove that globalization increases inequality, whereas numerous other studies claim that globalization reduces inequality.

The neoliberal argument says that the distribution of income between all of world’s people has become more equal over the past two decades and the number of people living in extreme poverty has fallen, for the first time in more than a century and a half. It says that these progressive trends are due in large part to the rising density of economic integration between countries, which has made for rising efficiency of resource use worldwide as countries and regions specialize in line with their comparative advantage for instance, attributed changes in the widened income inequalities to trade expansion. Mundell suggested that FDI flows contribute to a reduction of income inequalities in developing countries. Feenstra and Hanson argued that FDI flows into developing countries cause a higher wage for skilled workers than unskilled workers, resulting in widened income inequalities. Figini and Gorg argued that, initially, wage inequality increases with the FDI inflows, but, as blue-collar workers become skilled, that decreases in turn. Feenstra and Hanson’s and Figini and Gorg’s works were restricted to the Mexican and Irish cases, respectively and they did not compare the explanatory powers of the competing hypotheses.

Continue reading

# The Distributive Effects of Joining the Global Economy in Iran: the Application of ARDL Model – Introduction

Globalization is defined as the free movements of goods, services and capital across borders. It is a contentious process by which the western market economies have effectively spread across the globe. Although it does not constitute a new phenomenon, it is viewed as an inexorable integration of markets, nations and technologies to a degree never witnessed before in a way that is enabling individuals, and corporations to reach around the world further, faster, deeper and more economically than ever before. By contrast, some groups of scholars and activists view globalization as an ideological project of economic liberalization that subjects states and individuals to more intense market forces.

Globalization is widely believed to have had a generally positive impact on global economic growth. But the effect of globalization on employment and the distribution of incomes have been intensely debated in recent years.

Continue reading