**4.6 Modified Wald test for Granger causality test (M(WALD))**

From **Table 8** we have the lnoilpr as the dependent variable, at 5% level of significance, we accept the null hypothesis that there is no causality between, the lnexchr, lncpi and lnintr on the dependent variable. Also, the combination of all the independent variables do not granger caused changes in the dependent variable. This indicates the exogeneity of oil price which is been determined by many factors that are exogenous to both net importers and exporters of oil, Nigerian inclusive. According to Humbatova and Hajiyev [24] posited that the determinants of oil price range from financial factors, lack of production capacities in oil production, the decline in the world oil reserves, natural disasters, political events and processes, and no one country has the monopoly of determining oil price.

From **Table 9** we have the lnexchr as the dependent variable, at 10% level of significance, we reject the null hypothesis that there is no causality between loilpr and lnexchr. The exchange rate plays a significant role in determining the oil price both to net exporters and net importers. Specifically, oil is priced in U.S. dollars. According to Farley [64] submissions, each decrease and increase in the dollar or the price of the commodity (oil) generates an instantaneous realignment between


*Source: Estimation was compiled using E-views Version 9 software. Note: significance at 10% and 5% levels of significance respectively.*

#### **Table 8.**

*Granger causality test WALD test for Eq. (2) for the dependent variable: LNOILPR.*


*Source: Estimation was compiled using E-views Version 9 software. Note: significance at 10% and 5% levels of significance respectively.*

#### **Table 9.**

*Granger causality test WALD test for Eq. (3) for the dependent variable: LNEXCHR.*

#### *Impact of Oil Price Fluctuation on the Economy of Nigeria, the Core Analysis for Energy… DOI: http://dx.doi.org/10.5772/intechopen.94055*

the US dollar and other currencies. These correlated is more significant in countries with significant oil reserves that depend largely on crude exports and they experience more economic damage than those with more diverse resources. In the presentations of Bützer [65], he established that oil Net exporters tend to respond against depreciation pressures by running down foreign exchange reserves, particularly after oil demand shocks, but also global demand shocks (which also decrease oil prices). This is sometimes supplemented by a nominal depreciation of exchange rates. These invariably indicate that oil demand shocks are a relevant factor for their exchange rates. While we accept the null hypothesis that there is no causality between, the lncpi and lnintr on the dependent variable. Also, the combination of all the independent variables do not Granger cause changes in the dependent variable.

Also from **Table 10** we have the lncpi as the dependent variable, at 10% level of significance, we reject the null hypothesis and accept the alternative hypothesis that there is causality from lnexchr and linintr to lncpi. Exchange rate plays a vital role in determining prices in Nigeria, as an economy that has some element of a Dutch disease syndrome, and relied heavily on importation of basic necessity, when we factor out oil exportation from the total export, the non-oil balance of trade approximately stood at negative 7114 billion for 2017 as stated in our introduction. Therefore, appreciation in the exchange rate can cause inflation (lncpi) (Katz, 1973). The interest rate is one of the instruments used by the monetary authority to regulate the economy either during inflation or deflationary periods, the interest rate affects the demand and allocation of the available loanable funds the level, and pattern of consumption and investment ([66] p. 15). Before 2016 recession in Nigeria, the inflation rate was at a single digit of 9.55% in 2015, during the recession, the inflation rate was at double-digit 18.55% in 2016 and the central bank introduced a tight monetary policy, by raising the interest rate steady at 14 per cent from July 2017 to the first quarter of 2018 against 2016 which is 200 points higher [36].

Also, the combination of all the independent variables (lnoilpr, lnexchr and lnintr) does Granger cause changes in the dependent variable lncpi at 5%, but lnexchr and lnintr are more pronounced in the causality. While we accept the null hypothesis that lnoilpr do not granger cause lncpi.

In **Table 11** we have lnintr as the dependent variable, we reject the null hypothesis and accept the alternative hypothesis that at 5% levels of significance that there is a causality which is from lnoilpr and lnexchr to the endogenous variable lnintr, while there is no any causality with the log of lncpi on the dependent variable. Also, the combination of all the independent variables Granger cause changes in the dependent variable at a 5% level of significance. The relationship of lnoilpr and lnintr may not be exclusive but via the exchange rate, in the boom period the net exporter of oil has more dollars to expend, vice versa during deflationary periods, both periods has a direct link to economic growth. To avoid these inflationary or


*Source: Estimation was compiled using E-views Version 9 software. Note: \* and \*\* show significance at 10% and 5% levels of significance.*

#### **Table 10.**

*Granger causality test WALD test for Eq. (4) for the dependent variable: LNCPI.*


*Source: Estimation was compiled using E-views Version 9 software. \* and \*\* show significance at 10%, 5% and 1% levels of significance.*

#### **Table 11.**

*Granger causality test WALD test for Eq. (5) for dependent variable: LNINTR.*

deflationary tendencies, the central bank may engage in the sterilization process through open market operation, by manipulating the short-term interest rate, that is by increasing interest rates to discourage borrowing during inflationary periods or decrease the interest rate to encourage borrowing during deflationary periods. The relation is said to be inverse and this shows how oil price and exchange rate influences the monetary policy of net oil exporters.

#### **4.7 Forecast error variance decomposition (FEVD) and impulse response functions (IRFs)**

From the estimated TY-VAR, we compute forecast error variance decompositions (FEVD and impulse response functions (IRF), which serve as means for evaluating the dynamics of the interrelationship, interactions, and strength of causal relations among the variables in the system. The impulse response functions trace the effects of a shock to one endogenous variable on to the other variables in the VAR, variance decomposition separates the variation in an endogenous variable into the component shocks to the VAR [10, 46].

In simulating FEVD and IFRs, the VAR innovations can be contemporaneously correlated. That is a shock in one variable can work through the contemporaneous correlation with innovations in other variables. The responses of a variable to innovations in another variable of interest cannot be adequately represented in isolation, due to the facts that shock to individual variables cannot be separately identified due to contemporaneous correlation [46].

In our analyses, we applied Cholesky approach which uses the inverse of the Cholesky factor of the residual covariance matrix to orthogonalise impulses (innovations) as recommended by Sims (1980) as quoted by Duasa [46] and (Breitung, Bruggemann, and [58]) to solve this identification problem. The strategy requires a pre-specified causal ordering of the variables, which we estimated in **Table 5** for the correlation matrix. The results of FEVD are displayed in **Tables 12**–**15**, while the IRFs represented in **Figures 2**–**17** in appendix 1, respectively.

#### *4.7.1 Forecast error variance decomposition (FEVD)*

We explored the Cholesky factorization in the E-Views software and forecast the interrelationship of the variables up 48 months equal to 4 years. **Table 10** is the Table for FEVD for lnoilpr as a dependent variable for 48 periods (4 years) forecast. In forecasting a variable, shocks in the residual of the forecasted variable contribute more to its variance than the shocks in other variables in the first period. The shocks in oil price-output contributed more to its variance, from 100% in the first period down to 70.58% in the 48 period (4th year) of the forecast period. This is followed


*Impact of Oil Price Fluctuation on the Economy of Nigeria, the Core Analysis for Energy… DOI: http://dx.doi.org/10.5772/intechopen.94055*

*Note: SE refers to the total variance error in forecasting LNOILPR. Other columns represent the percentage of the variance attributable to shocks in the residual of the respective variables.* Sources: *Compiled using Eviews version 9.*

#### **Table 12.**

*Variance decomposition of LNOILPR.*

by lnintr that contributed 4.11% in the 24th period to about 18.11% in the 48 period (4th year). This followed by lncpi that contributed 1.48% at the 24th period to 7.09 at the 48 periods and last is the lnexchr contributions from 0.06% in the 24th period to 4.22% in the 48 periods. This shows monetary policy influences the fluctuation inherent with the oil price and in the future, it shows that lnintr will respond highly to oil price shocks. While the contemporaneous relationship between the oil prices as the endogenous variables (lncpi and lnexchr) in our model are very insignificant. This is an indication that it will take a longer time into the future, for variables other than lnintr to influence the impact of oil prices.

**Table 13**, is the Variance Decomposition for dependent variable lnexchr, the contributions to itself were 97.56% in the 1st period, to about 57.82% in the 48 period (4th year) into the future. This followed by the contributions of lnoilpr with 28.28% at the 24th period and 39.31% at the 48th period. While lncpi and lnintr contributed 2.58% and 0.02% all at the 48th period. The error variance in forecasting lnexchr from lnoilpr is high, which indicates that shocks in the residuals of lnoilpr will have much effect in determining the lnexchr in the future.

**Table 14** is forecast error variance decomposition of LNCPI as the predictant, the predictant contributes 99.81%, 54.73%, 3.18% in the 1st, 12th and 48th periods to itself, which indicates that the contributions of lncpi to itself declined in 4 years. While lnexchr contributes more to the error variance in forecasting lncpi, contributing about 43.40% up to 82.74% for the periods 12th and 36th then declined to 71.74%in the 48th period (4th year). While lnoilpr contributions started from 24th period with 2.47% and keep increasing up to 25.02% in the 48th period. Whereas


*Note: SE refers to the total variance error in forecasting LNEXCHR. Other columns represent the percentage of the variance attributable to shocks in the residual of the respective variables. Source: Estimation was compiled using E-views Version 9 software.*

#### **Table 13.**

*Variance decomposition of LNEXCHR.*

lnintr contributions are insignificant. This has brought a clearer picture that lnexchr and lnoilpr are the major determinant of inflation in the economy.

**Table 15** illustrated the forecast error variance decomposition of lnintr, contributing to its future error variation of 97.41%, 42.01% and 54.34% for the 1st, 12th and declined to 3.70% at the 48th period (4th year), this is followed by lnexchr which contributes 1.91%, 10.19% for the 1st and 6th periods, it declined for some periods and pick up again and continue rising to 82.81% in the 48th period (4th year).

This is trailed behind by lnoilpr, contributing 4.32% and 43.37% in the 6th and 12th, 75.25% at 24th period and started declining up to 12.41% at the 48th period (4th year). This indicates also a strong relationship into the future. The forecast error variance decomposition of the variables estimates also coincides with the result we obtained in the estimates we derived in **Table 11**, which also indicates that our estimates are good to go with for future implementation of policies.

#### *4.7.2 Response functions (IRFs)*

In **Figure 2**, from appendix 1, the Oil price (lnoilp) responded contemporaneously by the change in its own shocks, which is positive and not dissipating. The implication is that hick in the price of oil may mean high revenue, but the


*Impact of Oil Price Fluctuation on the Economy of Nigeria, the Core Analysis for Energy… DOI: http://dx.doi.org/10.5772/intechopen.94055*

*Note: SE refers to the total variance error in forecasting LNCPI. Other columns represent the percentage of the variance attributable to shocks in the residual of the respective variables. Source: Estimation was compiled using E-views Version 9 software.*

#### **Table 14.**

*Variance decomposition of LNCPI.*

consequences is, as an import based economic of non-oil goods and refined petroleum product, with domestic regulation of prices (subsidies), the policy will confine government's ability to finance the import bills as well as meet other international obligations [8]. While the response of oil price (lnoilpr) to change in Exchange rate (lnexchr) is insignificant in **Figure 3**. Inflation (lncpi), and Interest rate (lnintr) in **Figures 4**, and **5** showed some level of positive response.

In **Figure 6**, there is a slightly positive response of Exchange (lnexchr) to change Oil price (lnoilpr) in the sixth lag period. This show how influential oil is in determining exchange rate, since high price of oil means more revenue (foreign income), also Exchange (lnexchr) responded instantaneously, a positive response, to change in its self (**Figure 7**.). In **Figure 8**, there is slight positive response of lnexchr to change in lncpi and **Figure 9** showed a small inverse response of lnexchr to change in lnintr.

In **Figures 10** and **13**, Inflation (lncpi) did not show a meaningful response to orthogonal change in the price of oil (lnoilpr) and Interest rate (lnintr). While **Figure 11**, showed a positive response in Inflation (lncpi) to change in the Exchange rate (lnexchr), that is from the second lag period up to the tenth lag period in increasing order, this indicate that inflation will continue since the response is not dissipating unless there is a policy to induce deflation. Whereas in **Figure 12** there is an instantaneous response of Inflation (lncpi) to change in Inflation (lncpi) in a


*Cholesky Ordering: LNOILPR LNEXCHR LNCPI LNINTR. Note: SE refers to the total variance error in forecasting LNINTR. Other columns represent the percentage of the variance attributable to shocks in the residual of the respective variables. Source: Estimation was compiled using E-views Version 9 software.*

#### **Table 15.**

*Variance decomposition of LNINTR.*

#### **Figure 2.**

*Impulse response function of lnoilpr to lnoilpr.*

high positive level, with a slight drop towards the tenth period which indicates tendencies of achieving normality in the future.

**Figure 14**, showed that there is an inverse response of Interest rate (lnintr) to one standard deviation change in the price of oil (lnoilpr) from the second lag period in an increasing order up to the tenth period, this is expected because the assumption is that interest rate has an inverse relationship with the oil price. Also *Impact of Oil Price Fluctuation on the Economy of Nigeria, the Core Analysis for Energy… DOI: http://dx.doi.org/10.5772/intechopen.94055*

**Figure 3.** *Impulse response function of lnoilpr to lnexchr.*

#### **Figure 4.**

*Impulse response function of lnoilpr to lncpi.*

**Figure 5.** *Impulse response function of lnoilpr to lnintr.*

**Figure 15** indicated an instantaneous positive response of interest rate (lnintr) to change in the Exchange rate (lnexchr), in the third and fourth period, before it dying off which indicates that there is propensities of achieving normality in the long run. In **Figure 16** Interest rate (lnintr) responds contemporaneously to change in Inflation (lncpi), with a positive increase from the fourth period and finally, in **Figure 17** Inflation (lncpi) responded significantly to change Inflation (lncpi). The impulse response functions further complement the Forecast Error Variance Decomposition by given a portrait of the direction of the inter-relationships of variables.

**Figure 6.**

*Impulse response function of lnexchr to lnoilpr.*

**Figure 7.**

*Impulse response function of lnexchr to lnexchr.*

**Figure 8.** *Impulse response function of lnexchr to lncpi.*

## **5. Conclusion and recommendation**

In this research work, we explored the Toda-Yamamoto Modified Wald Test (MWALD) to examine the impact of oil price fluctuation on the monetary instrument in Nigeria, by looking at their causal relationships. The study covered the period 1995 to 2018 and the data are monthly data, to establish the contemporaneous relationships between these macroeconomic indicators. Among other analyses are the Granger Causality, FEVD and IRFs.

The review showed the direction of causality and FEVD into the future for 48 months equivalent to four years (short-run), between oil price, Exchange rate, Inflation, and Interest rate.

*Impact of Oil Price Fluctuation on the Economy of Nigeria, the Core Analysis for Energy… DOI: http://dx.doi.org/10.5772/intechopen.94055*

**Figure 9.**

*Impulse response function of lnexchr to lnintr.*

**Figure 10.**

*Impulse response function of lncpi to lnoilpr.*

**Figure 11.**

*Impulse response function of lncpi to lnexchr.*

From the analyses of Toda-Yamamoto Granger Causality WALD Test, the review presented that there is unidirectional causality from lnoilpr to lnexchr in **Table 9**. This is consistence with the result we obtained in the estimation forecast error variance decomposition of lnexchr (**Table 13**) as the predictant, where the predictant contributes 97.56% in the 1st period, to about 57.82% in the 48 period (4th year) into the future. This was followed by the contributions of lnoilpr with 28.28% at the 24th period and 39.31% at the 48th period. While lncpi and lnintr contributed 2.58% and 0.02% all at the 48th period. This was also complemented by for IRFs in **Figure 7** in the appendix.

Also from granger causality of lncpi as a dependent variable in **Table 10** there is unidirectional causality from lnexchr and lnintr to lncpi, also the combination of all

**Figure 12.**

*Impulse response function of lncpi to lncpi.*

#### **Figure 13.**

*Impulse response function of lncpi to lnintr.*

**Figure 14.**

*Impulse response function of lnintr to lnoilpr.*

the three independent variables (lnoilpr, lnexchr and lnintr) granger cause lncpi but lnexchr and lnintr have more contributions. This is also in tandem with the result of FEVD for dependent variable lncpi in **Table 14** where the dependent variable contributions to itself were 99.81%, 54.73%, 3.18% in the 1st, 12th and 48th periods, which indicates that the contributions of lncpi to itself declined in 4 years. While lnexchr contributes more to the error variance in forecasting lncpi, contributing about 43.40% up to 82.74% for the periods 12th and 36th periods (3rd years) then declined to 71.74%in the 48th period (4th year). While lnoilpr contributions started *Impact of Oil Price Fluctuation on the Economy of Nigeria, the Core Analysis for Energy… DOI: http://dx.doi.org/10.5772/intechopen.94055*

**Figure 15.**

*Impulse response function of lnintr to lnexchr.*

#### **Figure 16.**

*Impulse response function of lnintr to lncpi.*

**Figure 17.**

*Impulse response function of lnintr to lnintr.*

from 24th period with 2.47% and keep increasing up to 25.02% in the 48th period (4th year). This is also affirmed in **Figure 11** in the appendix.

Similarly in the estimation of Granger Causality WALD Test for lnintr, it responded positively to change in lnoilpr and lnexchr. This is also in agreement with the estimation of forecast error variance decomposition of lnintr as an endogenous variable, contributing to its future error variation of 97.41%, 42.01% and 54.34% for the 1st, 12th periods and declined to 3.70% at the 48th period (4th year), this is followed by lnexchr which contributes 1.91%, 10.19% for the 1st and 6th perods, it declined for some periods and pick up again and continue rising to 82.81% in the

48th period (4th year). This is trailed behind by lnoilpr, contributing 4.32% and 43.37% in the 6th and 12th, 75.25% at 24th period and started declining up to 12.41% at the 48th period (4th year). This indicated that the major determinant factors of interest rate policy in Nigeria are change in price of oil and exchange rate in the long run. This also conforms to the outcome of the IRF in **Figure 14**, which specified further that the relation between lnintr and lnoilpr is an inverse relationship, while lnexchr, lncpi and lnintr in **Figures 15**–**17** are positive.

The object of this is work is to establish a direct link between oil price and some selected monetary instruments in Nigeria, and our a priori expectations were achieved, we were able to established that oil price has a direct influence on the exchange rate, interest rate and inflation rate. It is known facts that Nigeria is an oilproducing economy and at the same time also an import-based economy of non-oil products. The major sources of financing the import come from oil revenue. As an oil-producing economy, there are tendencies of having Dutch disease syndrome and economic pass-through [9]. Both in theory and empirical analyses one can conclude that oil price is a strong determining factor of the rate of exchange, it has a direct link to inflationary or deflationary tendencies and also influences the monetary policies in Nigeria in terms of cost of borrowing.

Therefore, in implementation of monetary policy by the policymakers, attention should be drawn to price level of import from the external market, that is by concurrently monitoring the domestic market and the economy of the country's trading partners. On a general note, there should be diversification of the economy from oil to the non-oil economy to avoid the Dutch disease syndrome.

### **Additional classifications**

**JEL classifications**: *Q1, Q3, Q41, Q47*

## **Author details**

Jelilov Gylych\*, Abdullahi Ahmad Jibrin, Bilal Celik and Abdurrahman Isik Department of Economics, Nile University of Nigeria, Nigeria

\*Address all correspondence to: jelilov@nileuniversity.edu.ng

© 2020 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

*Impact of Oil Price Fluctuation on the Economy of Nigeria, the Core Analysis for Energy… DOI: http://dx.doi.org/10.5772/intechopen.94055*
