Online Access: www.absronline.org/journals

Management and Administrative Sciences Review Volume 4, Issue 2 Pages: 349-358 March, 2015

e-ISSN: 2308-1368 p-ISSN: 2310-872X

Examining the Dynamic Effects of Money Supply, Exchange Rate and Inflation in Iran Parvaneh Kamali1, and Sajede HasanNejadNeysi2 1. Associate Professor, PNU University of SharKord, Iran 2. Suasngerd Branch, Islamic Azad University, Susangerd, Iran

Given the importance of important economic variables because of affecting nature on the economic system, examining the relationship between exchange rate and money supply and growth rate of GDP are some of the research priorities for research economists and economic policy makers. Fundamental variables such as inflation, as one of the major economic problems that makes some problems for the economy's productivity and economic exchanges and exchange rate in terms of the country's dependence on foreign exchange from the oil sale are such variables that analyzing their influence on each other is essential. In this article using VAR method, this subject has been discussed during the period 1357-1390. In this article analysis tools include the impulse response and variance decomposition. According to the results, money supply has most effect on inflation in Iran and using Johansen test it shows long-run equilibrium relationship between variables. Keywords: inflation, foreign exchange, money supply, VAR method INTRODUCTION Achieving economic growth and development, along with an increase in the level of employment, controlling inflation and adjusting balance of payments have always been the ultimate goal of, the economy (Jeanne, 2012). Therefore, government fiscal policy tools and central bank's monetary policies are some tools that are used by countries to achieve these goals. Specifically monetary policies in the area of macroeconomic goals, seek price stabilization, adjusting balance of payments and controlling money or liquidity (Olivera, 2014). In this regard, monetary policy makers should have a careful evaluation of their effects on the economy. Studying "monetary transmission mechanism"

could assist policy makers in this area. Monetary transmission mechanism introduces the influence channels through which it affects monetary policies of decisions of firms, households, financial intermediaries and investors and it is followed by a change of economic activities (Lutz, 2014). Monetary policy is effective on exchange rate through changes in volume of money and interest rates and a condition of granting financial facilities. The objective of monetary policies in advanced industrial countries and developing countries is somewhat different. In industrialized countries the goal is removing the inflation; eliminate slack and

* Corresponding author: Forough Pahlavani Nejad, Field of Study Educational Management, Islamic Azad University of Boroujerd ,Iran. E-Mail: [email protected]

349

Manag. Adm. Sci. Rev. e-ISSN: 2308-1368, p-ISSN: 2310-872X Volume: 4, Issue: 2, Pages: 349-358

achieving full employment (Frenkel & Johnson, 2013). While for developing countries, the major goal of monetary policy has been economic growth and increasing government revenues and total supply. This paper tries to examine the effects of exchange rate and GDP growth rate and money supply on inflation in Iran. THEORETICAL HISTORY

BASIS

AND

RESEARCH

Factors affecting exchange rates One of the important factors affecting the exchange rate is domestic inflation that fundamental variables such as money supply and production levels are important for it. Inflation developments show their effects, sooner or later on the exchange rate over time. On the other hand, the country's trade balance is crucial (Ehrmann, Fratzscher, & Rigobon, 2011). In this regard, one of the complexities in determining the exchange rate in Iran is the fact that the major supplier of foreign exchange is the state. In other words, most of the exchange supply in the foreign exchange market is from oil exports carried out by government as an exogenous variable and influenced by the global oil market (Jamal & Sundar, 2011). Thus, the country's trade balance that is heavily influenced by earnings from oil exports could affect the value of the exchange rate. It is complicated because if exchange rate system is define floating, the nominal exchange rate increase (or decrease) following the value of injected exchange by the state reduce (or increase), and the business structure will face constant fluctuations (Lutz, 2014). Volatile trade leads to the unstable production and unstable economic growth. Notes that the amount of injecting oil exchange is result of the government financial and budgetary policies that their important feature is continuing budget deficit (Evans & Lyons, 2012). On the other hand, if the exchange rate system is selected a fixed or even controlled system, we may encounter experience of 80 decade in which because of high power of central bank in intervenes in the foreign exchange market (in support of a positive trade balance of abundant oil revenues), exchange rate was determined apparently stable, but inconsistent with the other economic variables. 80 decade was an experience in which the exchange rate in a medium

term was determined lower than the long-run equilibrium level and because of this the currency market was severely threatened by exchange rate increase (Liu & Nagurney, 2011). Real exchange rate developments The ratio of foreign prices to the domestic prices, according to the same currency or real exchange rate is an important factor affecting the composition of exports and imports and finally structure of economic activities that in Iran economy it must be well managed by the enumerated characteristics (Liu & Nagurney, 2011). Recent currency crisis can be explained on the basis of developments in this respect. According to the chart of the real exchange rate from the first quarter of 69 to the third quarter of 91, the culmination can be shown from 74 (the year of currency debt crisis) and 80 (the year of homogenization the exchange rates). Years before each culmination are a period in which the real exchange rate has decreased (Berman, Martin, & Mayer, 2012). Reducing the real exchange rate means that the economy is more dependent on imports and exports encounter economic losses. Note that this index will peak when the nominal exchange rate growth is more than domestic inflation differentials of foreign inflation (Broda & Romalis, 2011). The increasing nominal exchange rate is to return the real exchange rate to its long-run equilibrium path. Why the declining trend of real exchange rate is not permanent? Because government intervention through monetary and real shocks applying that doesn’t diminish the full effect of fundamental variables on exchange rates cannot be permanent and thus creating new information about future market and changing expectations, the nominal exchange rate increase will occur to compensate the previous accumulated inflation (Rose, 2011). Research history Clarida & Gali in their study (1994) use the three variable models, including the rate of change in real product, real exchange rate and price level, to study the cause of fluctuation of the real exchange. In this study, the effect of real shocks on the supply, real shocks on the demand and real shocks on real exchange in the long-run have been studied. Their analysis is based on quarterly data (from the third quarter of 1974 to the fourth quarter of 1992) related to the four countries of Japan,

350

Manag. Adm. Sci. Rev. e-ISSN: 2308-1368, p-ISSN: 2310-872X Volume: 4, Issue: 2, Pages: 349-358

Germany, United Kingdom, Canada that all three variables have been considered corresponding to their variable in the US. The results of this study show that nominal shocks in Japan and Germany, explain the majority of the variance of real exchange rate. Supply shock has little effect on the real exchange rate fluctuations in all four countries and in other 2 countries demand shocks has had greater impact on exchange rate fluctuations. Cunado & Gracia (2005) in a study entitled, oil prices, economic activities and inflation, have studied the impact of oil price fluctuations on 6 Asian countries during the period (1975-2002). In this study, both cumulative and Granger causality test are used. The results show that there is causality from oil shocks to the inflation rate in Japan, Singapore and Thailand. The relation of causality of the oil shock to the economic growth is proven in Japan, South Korea and Thailand. In general, asymmetries results have been observed in relation to changes in oil prices and inflation rate in the countries covered in this study. They showed that oil prices had a significant effect on the consumer price index, and this effect is small in the short term. Edwards (2006) examines the relationship between exchange rate and inflation in 7 countries (2 developed countries and 5 newly industrializing countries that have developed policy of reducing production over the past 50 years). He uses quarterly data over the period 1985 to 2000, OLS in his analysis. The main results can be summarized as follows: 1- Effect of exchange rate changes on inflation has been declined in countries that have targeted inflation reduction. 2- Implementation of monetary policy of inflation control has not resulted in an increase in the nominal and real exchange rate volatility. 3- There has been some evidence show increase in exchange rate in countries experienced high inflation rates in previous years. Shahab (1376) examined the relationship between exchange and inflation rate in Iran in a study entitled "exchange and inflation rates: an empirical analysis in Iran". Using the error correction model, he has shown the Inflation of currency fluctuations (directly and indirectly) in the period of 1374-1347 and he has emphasized that exchange rate fluctuations in the consumer price index, the price

index adjustment of the GNP and GDP price index and producer price index have had the greatest effect (negative) respectively. In addition, this study showed that long-term convergence of price index with more informal exchange rate is more than this intensity with nominal exchange rate. Khoshbakht (1386) in a study entitled "Studying the process of exchange rate impact on inflation of consumer price index and the import of Iran" has studied the exchange rate changes to the consumer price index and import using structural vector regression model and response functions and variance analysis of Cholsky, seasonally during the period 1383-1369. The results of this study indicate that the transmission of exchange rate volatility on the import price index was higher than the consumer price index. This is consistent with relatively higher share of tradable goods in the import price index to the consumer price index. On the other hand, the rate of change transition is money supply to the import price index. Model specification This study focuses on four variables based on empirical analysis. The variables we used include: Index of money supply (M2), inflation index (based on the base year 1383) (CPI), the real exchange rate index (REER) (every dollar to Rial) and GDP growth rate (LRGDP). Time series data from 1355 to 1390 is considered. Relationships between variables are specified as follows: Lcpi=B0+B1LM2+B2LREER+B3RGDP

(1)

In which CPI is the inflation index. LM2 is money supply and LREER is real exchange rate and RGDP is GDP growth. Results of data description Static evaluation of variables of the model For the stability test of time series of variables in this study, before the estimating regression model, stability test is used for all-time series. If the time series were not stable due to spurious regression problem, there is no possibility of using Autoregressive models. For the test stability, the unit root tests are used. One of the most common tests for recognition unit root is generalized DickyFuller test; this test has been used in this study.

351

Manag. Adm. Sci. Rev. e-ISSN: 2308-1368, p-ISSN: 2310-872X Volume: 4, Issue: 2, Pages: 349-358

As indicated in Table (1-3) generalizations are instable in 99% level for all model variables based on Dicky- Fuller unit root test model. TABLE 1 HERE But after subtraction, in the first difference they had stability that the logarithm of the exchange rate and the logarithm of GDP growth rate were with the logarithm and without stable trend. Generalized Dicky- Fuller unit root test results have been shown in table 2-3. TABLE 2 HERE The first step for estimating a vector error correction model is to determine the appropriate interval for the difference between the variables in the model. In Johanson method estimates of longterm relationships are very sensitive to the selected interval length. Furthermore, if many intervals are selected, high degree of freedom is lost and if intervals are selected less than the ideal amount interference sentences of equations will be correlated, so the optimal interval is important.

parameters are 34.46 and 24.26, respectively, which are larger than the critical values. The basis of selecting the model in this paper suggest that if there is a long term relationship it is needed to estimate vector error correction model. According to Johansen test results confirmed the presence of a long term relationship between the variables of the model, it was necessary to be taken in this regard by the error correction model. TABLE 4 HERE Results of estimating vector error correction model Long-run relationship between some variables is linear combinations between them, which is not unique. In software Eviews.6, the estimated longrun relationship has been normal based on one of the variables and long-term relationship has been normal based on (LCPI). In obtained long-term relationships, the vector error correction model studies the net impact of each of the explanatory variables on the dependent variable separately. Estimated long-term vectors

TABLE 3 HERE FIGURE 1 HERE As results of above table shows, the SBC, AIC, HQ, in comparison with other values, are minimum at first interval. In this model the optimal interval is selected one. To select the optimal number of intervals in the model, Schwartz and Akaike criterion is used, because this measure based on the principle of Persian Monis, suggests fewer intervals and finally, it provides a frugal model. In general, all indicators introduce interval as an optimal VAR interval. Co-integration test The results show that the hypothesis of lack of long-term relationship in the model is rejected, because the statistical value of the effect that is 81.67 is greater than its critical value at 5% level that is 47.85 and the maximum eigenvalue that is 47.20 is greater than the critical value that is 27.58 (and the hypothesis that there is no long-run relationship cannot be accepted. According to the presented results, the hypothesis of the existence of a long-term relationship is rejected at 5% significance level, because the values of these two

LCPI=19.26 LM2- 3.59LREER + 0.50 LRGDP (2) t:

(6.39)

(3.07)

(1.44)

As equation (2) shows, money supply has the most effect on inflation and then the real exchange rate has more effect and finally GDP growth rate has slight effect on inflation. Studying short-term relationship between variables in vector error correction model According to Angel-Granger (1978) theory, there should be a short-term relationship as error correction mechanisms to ensure the long-run equilibrium corresponding to the long-term economic relationship. In this section meanwhile studying the short term relationship between variables, using the error correction model of shortterm fluctuations of variables are related to the value of long-term variables. Summary of results of the short term mechanisms related to long-term model (6-3) is presented in the following table. TABLE 6 HERE As seen in the short term relationship, there is a relationship between money supply and real

352

Manag. Adm. Sci. Rev. e-ISSN: 2308-1368, p-ISSN: 2310-872X Volume: 4, Issue: 2, Pages: 349-358

exchange rate and GDP growth rate. Table (6-3) have been determined that, Dynamic estimating and analyzing the model Reaction functions stimulation (IRFS) Reaction stimulation functions show variables dynamic behavior of the system during the time when the momentum is in the size of a standard deviation. In this section, the dynamic response of the model variables because of momentum in the size of a standard deviation based on the Cholski analysis LCPI has been provided for 10 periods as follows: TABLE 7 HERE FIGURE 2 HERE In Figure 1, the graph on the horizontal axis shows time on an annual basis and on the vertical axis is the percentage change. Inflation rate momentum has the most effects on itself and then on the growth rate of GDP, after that on the money supply and the real exchange rate. Momentum of inflation rate in Iran has the most effects in the short-term on the growth rate of GDP.

is the real exchange rate. Therefore, government should use some policies to reduce the money supply in the country. Suggestions to policy makers include: It is recommended that a crisis management program in a finite time horizon and with an emphasis on producing rescue should be developed for a short time. The long term goals shouldn’t be the main problems in resolving the short term crisis. Banking facilities in this transition period must be spent to save production. So if you think boycott is in the transition period so a crisis management plan should be written for this period. The basic strategy for long-term stability of exchange rates and prices of other financial assets is reducing inflation associated with the development of the business environment in order to reduce the potential threat of liquidity. The gradual decrease of dominance of supply of oil in the currency market in the long run and preventing its fluctuations in the short term with reviving the reserve account for the main purpose can be determining intensely.

ANOVA

REFERENCES

While the impulse response functions depict an internal variable on the other variables in the VAR model analysis of variance, separates changes in endogenous variables from other endogenous variable impulses therefore, analysis of variance provides data on the relative importance of each of the random impulses to influence the variables in the model.

Berman, N., Martin, P., & Mayer, T. (2012). How do different exporters react to exchange rate changes? The Quarterly Journal of Economics, 127(1), 437-492.

TABLE 8 HERE The results of analysis of variance test of forecast error are expressed as follows: the most fluctuations in the momentum of inflation rate is explained by momentum of inflation rate and then the momentum of rate of GDP growth and money supply and exchange rate are real. CONCLUSION AND SUGGESTIONS This paper examines the effect of dynamic real exchange rate and GDP growth rate and money supply on inflation rate. As a long-term relationship showed the factor with highest effect on the inflation is the money supply and after that

Broda, C., & Romalis, J. (2011). Identifying the relationship between trade and exchange rate volatility. Paper presented at the Commodity Prices and Markets, East Asia Seminar on Economics, Volume 20. Clarida, R., & Gali, J. (1994, December). Sources of real exchange-rate fluctuations: How important are nominal shocks?. In CarnegieRochester conference series on public policy (Vol. 41, pp. 1-56). North-Holland. Cunado, J., & De Gracia, F. P. (2005). Oil prices, economic activity and inflation: evidence for some Asian countries. The Quarterly Review of Economics and Finance, 45(1), 65-83. Edwards, S. (2006). The relationship between exchange rates and inflation targeting revisited (No. w12163). National Bureau of Economic Research.

353

Manag. Adm. Sci. Rev. e-ISSN: 2308-1368, p-ISSN: 2310-872X Volume: 4, Issue: 2, Pages: 349-358

Ehrmann, M., Fratzscher, M., & Rigobon, R. (2011). Stocks, bonds, money markets and exchange rates: measuring international financial transmission. Journal of Applied Econometrics, 26(6), 948-974. Evans, M. D., & Lyons, R. K. (2012). Exchange rate fundamentals and order flow. The Quarterly Journal of Finance, 2(04). Frenkel, J. A., & Johnson, H. G. (2013). The economics of exchange rates: Selected studies: Routledge. Jamal, A., & Sundar, C. (2011). Modeling exchange rates with neural networks. Journal of Applied Business Research (JABR), 14(1), 1-6.

Liu, Z., & Nagurney, A. (2011). Supply chain outsourcing under exchange rate risk and competition. Omega, 39(5), 539-549. Lutz, F. A. (2014). The case for flexible exchange rates. PSL Quarterly Review, 7(31). Olivera, J. H. (2014). Money, prices and fiscal lags: a note on the dynamics of inflation. PSL Quarterly Review, 20(82). Rose, A. K. (2011). " Exchange Rate Regimes in the Modern Era": Fixed, Floating, and Flaky. Journal of Economic Literature, 652-672. Shahab, Mojtaba, 1376, exchange and inflation rates: an empirical analysis about Iran. MSc Thesis, Department of Teacher Education.

Jeanne, O. (2012). Capital account policies and the real exchange rate: National Bureau of Economic Research. Khoshbakht, Amene, 1386, reviewing the impact of affecting exchange rate on inflation in the consumer price index and import in Iran, Economic Bulletin, No. 27: 51-82.

APPENDIX Table (1-3): Generalized Dicky- Fuller unit root test Results (research results) Critical Critical The values values Variables calculated number %1 %5 Logarithm of the real LREER (0) -1.51 -3.63 -2.95 exchange rate Logarithm of LCPI(0) -0.69 -3.64 -2.95 inflation Logarithm of the LM2(0) -1.86 -3.63 -2.95 money supply Logarithm of the rate LRGDP(0) 0.88 -3.63 -2.95 of growth of GDP

354

Manag. Adm. Sci. Rev. e-ISSN: 2308-1368, p-ISSN: 2310-872X Volume: 4, Issue: 2, Pages: 349-358

Table 2-3: Generalized Dicky- Fuller unit root test results with a degree of difference The Critical Critical values values Variables calculated number %1 %5 Logarithm of the real LREER (1) -4.2 -3.63 -2.95 exchange rate Logarithm of inflation LCPI(1) -4.43 -3.64 -2.95 Logarithm of the money supply Logarithm of the rate of growth of GDP

LM2(1)

-4.53

-3.63

-2.95

LRGDP(1)

-3.50

-3.63

-2.95

Table 3-3: Results of optimal interval selection

Table 4-3: cointegration test based on the work Zero hypothesis The test statistic result Critical value at 5% The lack of cointegration 81.67 47.85 There is 1 cointegration relationship 34.46 29.79 There are 2 cointegration relationship 10.19 15.49 Table 5-3: cointegration test based on the maximum eigenvalue test Zero hypothesis The test statistic result Critical value at 5% 47.20 27.58 The lack of cointegration 24.26 21.13 There is 1 cointegration relationship 9.43 14.26 There are 2 cointegration relationship

355

Manag. Adm. Sci. Rev. e-ISSN: 2308-1368, p-ISSN: 2310-872X Volume: 4, Issue: 2, Pages: 349-358 Table 6-3: results of vector estimating of Auto Regression Vector Error Correction Estimates Date: 11/18/14 Time: 19:20 Sample (adjusted): 1357 1390 Included observations: 34 after adjustments Standard errors in ( ) & t-statistics in [ ] CointEq1

Cointegrating Eq:

1.000000

LCPI(-1)

19.26660 (3.01375) [ 6.39291]

LM2(-1)

-3.599698 (1.17102) [-3.07398]

LREER(-1)

0.500318 (0.34655) [ 1.44370]

LRGDP(-1)

-64.13018

C

D(LRGDP)

D(LREER)

D(LM2)

D(LCPI)

Error Correction:

0.017029 (0.00321) [ 5.30391]

-0.004576 (0.01333) [-0.34324]

-0.042657 (0.01371) [-3.11188]

-0.002331 (0.00307) [-0.75975]

CointEq1

-0.091519 (0.17677) [-0.51774]

-0.207924 (0.73395) [-0.28330]

0.085701 (0.75468) [ 0.11356]

0.331722 (0.16893) [ 1.96372]

D(LCPI(-1))

-0.165788 (0.06151) [-2.69518]

0.367693 (0.25540) [ 1.43966]

0.702487 (0.26262) [ 2.67495]

0.111538 (0.05878) [ 1.89744]

D(LM2(-1))

-0.176149 (0.04179) [-4.21558]

0.202381 (0.17349) [ 1.16650]

-0.010826 (0.17839) [-0.06069]

0.009029 (0.03993) [ 0.22611]

D(LREER(-1))

-0.126097 (0.15685) [-0.80392]

-0.271436 (0.65126) [-0.41679]

1.006498 (0.66965) [ 1.50301]

0.258219 (0.14989) [ 1.72268]

D(LRGDP(-1))

0.218776 (0.03806) [ 5.74843]

0.096113 (0.15802) [ 0.60823]

-0.220224 (0.16248) [-1.35536]

0.068839 (0.03637) [ 1.89274]

C

0.694780 0.640277 0.116297 0.064447 12.74745 48.28155 -2.487150 -2.217792 0.181550 0.107454

0.233726 0.096892 2.004892 0.267588 1.708093 -0.120815 0.360048 0.629406 0.007726 0.281577

0.282553 0.154437 2.119746 0.275146 2.205451 -1.067814 0.415754 0.685111 -0.036437 0.299219

0.311425 0.188465 0.106206 0.061588 2.532741 49.82454 -2.577914 -2.308556 0.170454 0.068366

R-squared Adj. R-squared Sum sq. resids S.E. equation F-statistic Log likelihood Akaike AIC Schwarz SC Mean dependent S.D. dependent

356

Manag. Adm. Sci. Rev. e-ISSN: 2308-1368, p-ISSN: 2310-872X Volume: 4, Issue: 2, Pages: 349-358 5.81E-08 2.67E-08 103.4525 -4.438381 -3.181378

Determinant resid covariance (dof adj.) Determinant resid covariance Log likelihood Akaike information criterion Schwarz criterion

Table 7 reaction function stimulation in the size of a standard deviation in LCPI LRGDP LREER LM2 LCPI Period 0.000000 0.013707 0.019794 0.021530 0.021054 0.021180 0.022167 0.023195 0.023701 0.023777

0.000000 0.003002 -0.006696 -0.013526 -0.017171 -0.018549 -0.019580 -0.020821 -0.021970 -0.022699

0.000000 0.016781 0.020623 0.015440 0.013999 0.018154 0.023350 0.026051 0.026376 0.026090

0.061588 0.090073 0.099857 0.103190 0.103794 0.104320 0.105197 0.106016 0.106454 0.106583

LRGDP

Table 8-3: Analysis of variance LCPI LREER LM2 LCPI

0.000000 1.517134 2.496815 3.002045 3.199686 3.309973 3.419991 3.535906 3.638594 3.718508

0.000000 0.072787 0.231925 0.681393 1.144326 1.497867 1.774616 2.018841 2.244583 2.445186

0.000000 2.273849 3.044689 2.720247 2.456599 2.515867 2.841553 3.214188 3.498509 3.699937

100.0000 96.13623 94.22657 93.59632 93.19939 92.67629 91.96384 91.23106 90.61831 90.13637

1 2 3 4 5 6 7 8 9 10

S.E.

Period

0.061588 0.111287 0.152375 0.186417 0.215542 0.241792 0.266364 0.289550 0.311307 0.331713

1 2 3 4 5 6 7 8 9 10

357

Manag. Adm. Sci. Rev. e-ISSN: 2308-1368, p-ISSN: 2310-872X Volume: 4, Issue: 2, Pages: 349-358

Figure 1 the instantaneous reaction of inflation rate

Figure 2: Analysis of variance functions

358