See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/384637051

The Impact of Fixed and Floating Foreign Exchange Rates on Inflation Rates in
Nigeria: Statistical Evidence

Article · June 2025

CITATIONS READS

0 27

3 authors, including:

Enobong Francis Udoumoh

Federal University of Agriculture, Makurdi
26 PUBLICATIONS 43 CITATIONS

SEE PROFILE

All content following this page was uploaded by Enobong Francis Udoumoh on 05 October 2024.

The user has requested enhancement of the downloaded file.

Vol. 5 No. 1, June, 2025
Ikughur et al. FUAMJPAS 5(1):30-37 June, 2025

The Impact of Fixed and Floating Foreign Exchange Rates on Inflation

Rates in Nigeria: Statistical Evidence
J.A1*. IKUGHUR E.F.1 UDOUMOH and E.A2. ODUNEWU
1
Department of Statistics, Joseph Sarwuan Tarka University Makurdi, Nigeria
2
National Bureau of Statistics, Abuja

Corresponding author: [email protected]

Received: 24/07/2024 Accepted: 11/09/2024 Published online: 13/09/2024

Abstract
Most statistical applications to modeling and forecasting macroeconomic variables do attempt to provide the best model that
explains the relationship between these variables. In this study, application is performed with the view to provide explanation to
the impacts of the stable (fixed) and also unstable (floating) foreign exchange (Forex) market on inflation rate in Nigeria. The
stable Forex is represented by a stationary process while the unstable Forex market is represented by non-stationary process.
Using the Box-Jenkins Seasonal Autoregressive Integrated Moving Average with exogenous variable (SARIMAX) approach, the
results showed that fixed exchange rate market has the tendency to stabilize inflation in a developing economy as there would
be a reduction of approximately 50% in inflation rate in Nigeria when Forex is regulated, and that macroeconomic variables are
suitably predicted when they are generated through a stationary process.
Keyword: Stationary process, Inflation, Fixed forex, Floating forex, SARIMAX

Introduction and may eventually lead to an impairment of trust and

Inflation is a quantitative measure of the rate at which the confidence in a currency.
average price level of a basket of selected goods and Exchange Rate on the other hand is the rate at which the
services in an economy increases over a period of time. It currency of one country is being exchanged for that of
indicates a decrease in the purchasing power of a nation’s another country or the relative price that indicates the
currency [1]. As a critical economic variable, it interacts price of one currency in terms of another currency. [4]
jointly with many other factors, including economic defined it as the currency that can be exchanged as one
growth, employment, exchange rates, gross capital unit of another currency, or the value of currency with
formation and many more. Inflation is also seen as a another currency. [5] had earlier noted that when the
sustained rise in the average price of goods and services exchange rate is defined as the rate of change between
over time based on prices of goods that are commonly two national currencies increases, the overall level of
consumed by the public [2]. It is measured using the prices in the economy will also increase and when the
Consumer Price Index (CPI) which is the current price of exchange rate falls, the domestic currency appreciates, and
a collection of goods and services in terms of the same prices are expected to fall in the general level.
period prices in the previous year [3]. Similarly, [1] noted that if a country increases her
exposure to foreign marketplaces, its inflation will become
Consumer price index measures indicate the relative worse as it has impact on the cost of living, the cost of
movement in the general prices of goods and commodities doing business, borrowing money, mortgages, corporate
consumed by the public relative to a base period, being and government bond yields, and every other facet of the
made up of a basket of consumables. It is the general economy.
measure of the level of inflation existing in the economy According to [6], exchange rate changes can affect inflation
and its impact on the living standards of the people. rate through three different channels. Firstly, exchange
According to European Parliament briefing of 2021 rate changes experienced in an open economy affects the
(www.europarl.europa.eu) on the subject, inflation index is relative prices between domestic and foreign goods.
one of the most important economic indicators affecting Secondly, it affects the prices of imported final goods in
daily lives of people and organization. This is because, terms of national currencies and thirdly, it affects nominal
when prices increase significantly, the purchasing power of wages through the impact of imported middle prices in
individuals, households and firms’ declines and this makes terms of local currency.
it difficult for people and firms to plan to save or invest
Publication of College of Science, Joseph Sarwuan Tarka University, Makurdi
https://fuamjpas.org.ng/ 30
Ikughur et al. FUAMJPAS 5(1):30-37 June, 2025

According to [7], Nigeria has operated two exchange [18] studied with a large dataset for 71 countries, 52 of
regimes: fixed exchange regime from 1960 to 1986 and which were developing countries, covering the period
floating exchange rate regime from June 1986. Between 1979-2000. They examined the pass-through effect of
April 1974 and late 1976, the monetary authorities pegged inflation and found that it runs from exchange rates to
the naira to whichever currency was stronger in the domestic prices. Specifically, they drew attention to the
foreign exchange market, either the US dollar or the fact that the pass-through effect is subject to the inflation
British pound sterling ([8], [9]). Prior to the introduction regime and so, it is more significant in countries with
of the second-tier foreign exchange market, the higher inflation, purporting that the higher inflation, the
management of foreign exchange was through a fixed higher the pass-through effect.
exchange rate regime with the aim to ensure price [19] examined the extent to which changes in exchange
stability, which was necessary for imports of capital goods rates result in changes in Turkish domestic inflation in
for the development of the domestic economy. [10] order to determine the impact from pre-2003 period to
noted that adoption of floating exchange rate regime the post-2003, when the exchange rates were allowed to
followed the collapse of the fixed exchange regime and float. Employing monthly frequency data to estimate two
was formalized in January 1976 during the Jamaican impulse-response functions and pass-through coefficients,
meeting of the IMF (International Monetary Fund) they confirmed that exchange rate shocks feed into
members where rules for the International Monetary domestic inflation, first at the level of manufacturers’
system in place today were agreed. [11] noted that prices and then at the level of consumer prices.
although floating exchange rates does reflect the true value Again, [4] determined the effect of inflation on the value of
of a currency as against fixed rates which are arbitrary and rupiah in Indonesia in the Islamic finance. Evaluating the
tend to be favour developed nations than developing data using percentage change in these variables, the result
nations. showed that the influence of inflation toward currency
exchange occurred because of export decline thereby
In Nigeria, [12] reported on how Forex impact trade causing inflation to the value of the rupiah against US
noting that because the economy has been in a slump due dollar. Similarly, [20] investigated the relationship between
to inflation rate, among others, and also, the unsettling exchange rates and inflation in Iran. They used the Hendry
exchange rate which he sees as a major issue affecting General to Specific Modeling method and Vector
Nigeria’s economy because of the weak naira in relation s Autoregressive (VAR) model and concluded that as the
to Forex as Naira weakening can no longer be controlled. exchange rate increases, the inflation rate also increases.
This has caused increase in exchange expenses and prices [21] assessed the main drivers of inflation in Ethiopia and
which in turn, limits production capacity. By implication, Kenya by developing single-equation error correction
the fast-increasing prices in marketplace is seen to be models for the CPI in each country. The study took into
induces by the fast-increasing rate of foreign exchange account a number of potential sources of the recent surge
rate. in inflation, which included excess money supply, exchange
rates, food and non-food world prices, world energy
Earlier, [13] had analyzed the relationship between
prices and domestic agricultural supply shocks. Their
exchange rates and inflation and presented a theoretical
findings showed that inflation rates in both Ethiopia and
argument on the exchange rate−inflation nexus in the
Kenya were driven by similar factors including exchange
context of monetary policy credibility that, a fixed or
rates.
stable exchange rate policy could make easier the task of
the monetary authority in lowering inflation by increasing The study by [22] investigated the impact of inflation on
credibility. This argument was also upheld by [14] and [15] real exchange rate volatility in Nigeria, using quarterly 181
who contended that a stable exchange rate regime data items. They used the GARCH (1, 1) model and
provides not only price stability but also increase the Granger causality in Vector Auto-Regressive environment
efficiency of monetary policy. Similarly, [16] pointed out and the Granger causality tests. They showed that there is
that a fixed exchange rate regime increases the credibility a relationship between imported inflation, real exchange
of monetary policy, contribute to having lower rates of rate volatility and other sample variables and
inflation by creating a confidence effect on domestic recommended inflation targeting policy to control the
currency rather than foreign currencies fluctuation in the price level as well as other
macroeconomic variable that have a direct effect on the
However, in the floating exchange rate regime, unlike the
exchange rate
fixed one, there is overshooting of the equilibrium
Another study by [23] employed the Granger causality and
exchange rate in both directions and then, cause prices to
ARDL modeling approach on annual data from 1980 –
go up by increasing the domestic prices of imported goods
2021 to investigate the nexus between inflation and
when depreciation occurs. Thus, it can be claimed that
exchange rate in Nigeria. They found the existence of
inflation is likely to be higher under a floating exchange
cointegration relationship between inflation, exchange
rate regime in relation to a fixed one [13]. Furthermore,
rate, import and GDP as in the short run, exchange rate is
[17] argued that fixed exchange rates are less inflationary
found to have positive and statistically significant impact on
and the anti-inflationary benefit is heavily dependent on the
inflation while in the long run there exist a negative and
monetary stability and credibility of the regime, which
significance relationship between inflation and exchange
needs to be carefully built in a stable economy.
rate.
Publication of College of Science, Joseph Sarwuan Tarka University, Makurdi
https://fuamjpas.org.ng/
31
Ikughur et al. FUAMJPAS 5(1):30-37 June, 2025

Similarly, [24] investigated the impact of exchange rate [29] studied unemployment using the various unit root
volatility on inflation in Nigeria using the GARCH and tests noting that, if unemployment is defined as a unit root
VECM to ascertain the long run impact. The study found process, policymakers should focus on structural reforms
that nominal Exchange rate and money supply has positive to mitigate adverse shocks. On the contrary, if
impact on Consumer’s Price Index (CPI) indicating that unemployment is stationary, the goal would be to avoid
inflation in Nigeria is caused by exchange rate fluctuations short-term imbalances as unemployment hysteresis is
as well as increase in money supply. Furthermore, [25] associated with non-stationary unemployment rates. Thus,
examined the effect of exchange rate fluctuation on if there is a unit root in a series, it suggests that automatic
inflation rate in Nigeria using supporting control variables return to a normal trend may not occur.
such as interest rate, money supply, imports and growth of If a series does not contain a unit root, [30] observed that,
GDP. The OLS and GLS models were developed for data the underlying trend is deterministic and the series has a
analyses. The results showed that exchange rate and other short memory. Thus, the resulting shock has no
macroeconomic variables including interest rate, money permanent impact as the series does return to its steady
supply, imports and GDP do not impact on inflation in trend. Similarly, [31] had shown that a stationary process
Nigeria. has finite variance but a limited memory while a non-
stationary process has infinitely long memory and wanders
[11] in an empirical study of Nigeria’s foreign exchange widely. In essence, if Forex is stationarity, it provides
regimes opined that the present system of flexible indication of its stability in the exchange market.
exchange rates should be fine-tuned to allow for some The problem of modeling the impact and also, the
form of government intervention in fixing exchange rates relationship between Exchange rate and Inflation allowing
when the need arises. for fixed and floating exchange rates have not been
carefully addressed in the case of Nigeria’s economy which
is the target of this study. This study will also investigate
Non-Stationarity and Foreign Exchange Instability
the perception of operators in the open market on the
Non-stationarity occur when a certain attributes of a time
relationship between Naira and Foreign Exchange in US
series do vary with time. It is infact a condition where the
Dollar.
mean, variance and covariance of a time series change over
time thereby making such data not to possess stable Materials and Methods
behavior. Basically, stochastic time series models are based This study makes use of secondary data on inflation rate
on the assumptions that the forces exerting on the collected from National Bureau of Statistics (NBS) and
process generating such series will continue in similar exchange rate (US Dollar –Nigeria’s Naira) data from
manner. It is therefore, reasonable to employ stability Central Bank of Nigeria (CBN) statistical bulletin from
measures before modeling such variables. year (1996:1 to 2019:12). Data analysis The ARIMAX
A study by [26] pointed at the existence of unit roots in family modeling techniques is utilized in this study.
macroeconomic time series and supported it with ARIMAX Model
statistical evidence that the hypothesis of a unit root in the The ARIMAX model is an extension of Autoregressive
autoregressive representations of a dozen macroeconomic Integrated Moving Average (ARIMA) model which
time series existed for the United States. Furthermore, combined the AR(p) and the MA(q). The order of
[26] noted that when time series have a unit root, they are integration, d, represents the order of differencing in the
better characterized as non-stationary processes that have model. The ARIMAX model incorporates covariate(s) to
no tendency to return to a long-run deterministic path. the ARIMA model as follows:
Besides, the variance of the series is time-dependent and The ARIMAX model is of the form
goes to infinity as time approaches infinity, which results in
𝜙𝑝 (𝐵)(1 − 𝐵)𝑑 𝑦𝑡 = 𝜃𝑞 (𝐵)𝜀𝑡 + 𝛽𝑥𝑡 (1)
serious problems for forecasting. This is to say that non-
stationary series suffer permanent effects from random While the ARIMAX model taking cognizance of the
shocks. seasonal effect is given as
𝜙𝑝 (𝐵)Φ𝑃 (𝐵 𝑠 )(1 − 𝐵)𝑑 (1 − 𝐵 𝑠 )𝐷 𝑦𝑡 =
Unit root test, according to [27] helps to identify some 𝜃𝑞 (𝐵)Θ𝑄 (𝐵)(𝐵 𝑠 )𝜀𝑡 + 𝛽𝑥𝑡 (2)
features of the underlying data-generating process of a
series. According to [27], if a series has no unit roots, it is Where the characteristic polynomials 𝜙𝑝 , Φ𝑃 , 𝜃𝑞 , Θ𝑄 are
characterized as stationary, and therefore exhibits mean defined as follows:
reversion in that it fluctuates around a constant long run 𝜙(𝐵)=(1 − 𝜙1 𝐵 − 𝜙2 𝐵2 − ⋯ − 𝜙𝑝 𝐵𝑝 ). , 𝜃(𝐵)=(1 −
mean. It also implies that the series has a finite variance 𝜃1 𝐵 − 𝜃2 𝐵2 − ⋯ − 𝜃𝑝 𝐵𝑞 ).
which does not depend on time and that the effects of Φ𝑃 (𝐵)=(1 − Φ1 𝐵 𝑠 − Φ2 𝐵2𝑠 − ⋯ − Φ𝑝 𝐵𝑝𝑠 ). ,
shocks dissipate over time.
Another study of unit root as a method of detecting Θ𝑄 (𝐵)=(1 − Θ1 𝐵 𝑠 − Θ2 𝐵2𝑠 − ⋯ − Θ𝑄 𝐵𝑞𝑠 ).
financial bubble is due to [28] who examined several for the seasonal and non-seasonal AR(p), MA(q) where B
works on unit root methods as used in detection of is backshift 𝜀𝑡 is white noise, 𝑥𝑡 is a covariate at time t. In
financial bubbles in asset prices noting that fundamental this study, a single covariate is used whose coefficient is 𝛽.
changes in the autocorrelation structure of relevant time However, it is possible to have two or more covariates in
series imply the presence of a rational price bubble while the model. If s = 0 in Equation 2, we have the ARIMA
Publication of College of Science, Joseph Sarwuan Tarka University, Makurdi
https://fuamjpas.org.ng/ 32
Ikughur et al. FUAMJPAS 5(1):30-37 June, 2025

model specification while for s=0 and d=D = 0, we have Model Diagnostic
the ARMA model specified. The Jarque–Bera test is a goodness-of-fit test of whether
The SARIMAX process takes cognizance of exogenous sample data have the Skewness and kurtosis matching a
variables that are inherently seasonal in nature or can be normal distribution. If it is far from zero, it signals the data
established by any test of seasonality. Thus, Equation 2 can do not have a normal distribution. The test statistic JB is
also be described as SARIMAX (p,d,q) (P,D,Q)[S] process. defined as:
𝑛 1
𝐽𝐵 = (𝑆 2 + (𝐾 − 3)2 ) (6)
Test for Stationarity 6 4
Where n is the number of observations, S is the sample
The Dickey-Fuller test was the first statistical test
skewness computed as
developed to test for stationarity in a time series data. The 1 𝑛
𝜇
̂3 ∑ (𝑥 −𝑥)3
Augmented Dickey-Fuller test is given as 𝑆= = 𝑛 𝑖=1 𝑖
3⁄ ,
̂3
𝜎
∆𝑦𝑡 = 𝛼 + 𝛿𝑡 + 𝛽𝑦𝑡−1 + ∑𝑛𝑖=1 𝛾𝑖 ∆𝑦𝑡−𝑖 + 𝜀𝑡 (3) (
1
𝑛
∑𝑛 2
𝑖=1(𝑥𝑖 −𝑥) )
2

Under null hypothesis H0, 𝑦𝑡 is considered to be I(1) which K is the sample kurtosis given as
is equivalent to ∆𝑦𝑡 being I(0) in which case β would be 𝜇
̂4
1 𝑛
∑ (𝑥 −𝑥)4
𝑛 𝑖=1 𝑖
zero. The test statistic is the standard regression t-statistic 𝐾= = 2
̂4
𝜎 1 𝑛
( ∑𝑖=1(𝑥𝑖 −𝑥)2 )
is given by 𝑛

̂
𝛽
𝑡𝛽 = (4) If the data comes from a normal distribution, the JB
𝑆𝐸𝛽
̂
statistic asymptotically has a chi-squared distribution with
two degrees of freedom, so the statistic can be used to
Model Selection Criteria test the hypothesis that the data are from a normal
In choosing the model that best describes a time series distribution.
data, attention is given to the RMSE, AIC, BIC and AICc. The Ljung–Box test is a type of statistical test of whether
Smaller values indicate better model. any of a group of autocorrelations of a time series are
Define the log likelihood as different from zero. The null hypothesis is that the data
𝑆𝑆𝑄′ 2𝑁𝑙𝑛(2𝜋) are independently distributed against the alternative of
𝐿 = −𝑁𝑙𝑛(𝜎 2 𝑎 ) − − (5) dependence. The test statistic according to [32] is:
2𝜎 2 𝑎 2
The Akaike's Information Criterion (AIC) is given as ̂2
𝜌
𝑄 = 𝑛(𝑛 + 2) ∑ℎ𝑘=1 𝑘 (7)
𝑛−𝑘
𝐴𝐼𝐶 = −2𝐿 + 2𝑁𝑝
Where 𝑛 is the sample size, 𝜌̂𝑘 is the sample
While the Bayesian Information Criteria is 𝐵𝐼𝐶 = −2𝐿 + autocorrelation at lag k, and n is the number of lags being
ln (𝑁)𝑁𝑝 tested. Under H0 the statistic Q asymptotically follows a
Where N = Total number of observations, L= -2 log 2
𝜒(𝑛) . For significance level α, the critical region for
likelihood, 𝜎 2 𝑎 =variance of residuals 2
rejection of the hypothesis of randomness is 𝑄 > 𝜒1−α,𝑛
Np= Number of parameters(𝑁𝑝 = 𝑝 + 𝑞 + 𝑑 + 𝑃 + 𝑄 + 2
where 𝜒1−α,𝑛 is the (1 − α)-quantile of the chi-squared
𝐷 + 𝑚), SSQ = residuals sum of squares. distribution with h degrees of freedom [33].

Results and Discussions

The results in this study are shown on Figures 1-3 and Tables 1-4.
INFLATION
50

40

30

20

10

-10
96 98 00 02 04 06 08 10 12 14 16 18 20 22
Figure 1: Time plot of Nigeria’s Monthly Inflation Rates from 1996-2022

Publication of College of Science, Joseph Sarwuan Tarka University, Makurdi

https://fuamjpas.org.ng/ 33
Ikughur et al. FUAMJPAS 5(1):30-37 June, 2025

FOREX
450

400

350

300

250

200

150

100

50
96 98 00 02 04 06 08 10 12 14 16 18 20 22
Figure 2: Time plot of Nigeria’s Monthly Actual Foreign Exchange Rates from 1996-2022

D(FOREX)
120

80

40

-40

-80
96 98 00 02 04 06 08 10 12 14 16 18 20 22

Figure 3: Time plot of Nigeria’s FOREX from 1996 - 2022 at First Differences

Table 1: Unit Root Test for Inflation and Forex Time Series

Variable Levels tc t0.05 P Decision

Inflation I(0) -5.684 -3.426 0.000 Reject H0: Stationary at level of data

Forex I(0) -1.253 -3.426 0.8969 Do not Reject H0: Unit Root

D(Forex) I(1) -13.41 -3.426 0.000 Reject H0: No Unit Root

Table 2: Model Selection and Adequacy Criteria

Model Forex D(Forex)
Characteristics SARIMAX(1,0,1)(1,0,1)12 SARIMAX(1,0,1)(1,0,1)12
R2 0.914 0.914
SE-Regression 1.402 1.212
AIC 3.533 3.547
DW 1.985 1.982
RMSE 4.543 4.649
MAE 3.419 3.503
MAPE 226.178 232.082
BG Test (Serial Significant at 5% 0.1726 (P=0.8415
Publication of College of Science, Joseph Sarwuan Tarka University, Makurdi
https://fuamjpas.org.ng/ 34
Ikughur et al. FUAMJPAS 5(1):30-37 June, 2025

Correlation)
BP Test F=13.248, (P =0003) 2.746 (P= 0.100)
(Heteroskedasticity)

Table 3: SARIMAX (1,0,1)(1,0,1)12 Model with Forex as exogenous variable

Variable Coefficient Std. Error t-Statistic Prob.

C 10.58998 0.803157 13.18545 0.0000

FOREX 0.008054 0.003891 2.069748 0.0393
AR(1) 0.935391 0.021977 42.56263 0.0000
SAR(12) 0.208339 0.056990 3.655699 0.0003
MA(1) 0.173417 0.059138 2.932422 0.0036
SMA(12) -0.952707 0.009937 -95.87743 0.0000

R-squared 0.914432 Mean dependent var 11.96080

Adjusted R-squared 0.913029 S.D. dependent var 4.755661
S.E. of regression 1.402486 Akaike info criterion 3.533474
Sum squared resid 599.9250 Schwarz criterion 3.605624
Log likelihood -543.4552 Hannan-Quinn criter. 3.562313
F-statistic 651.8800 Durbin-Watson stat 1.985122
Prob(F-statistic) 0.000000

Table 4: SARIMAX (1,1,1)(1,0,1)12 Model with D(Forex) as exogenous variable

Variable Coefficient Std. Error t-Statistic Prob.

D(FOREX) 0.004233 0.004796 0.882743 0.3781

C 12.19193 0.322071 37.85478 0.0000
AR(1) 0.946529 0.020949 45.18346 0.0000
SAR(12) 0.208075 0.057379 3.626319 0.0003
MA(1) 0.171483 0.059096 2.901760 0.0040
SMA(12) -0.950645 0.010522 -90.34939 0.0000

R-squared 0.913557 Mean dependent var 11.95632

Adjusted R-squared 0.912135 S.D. dependent var 4.762693
S.E. of regression 1.411757 Akaike info criterion 3.546712
Sum squared resid 605.8896 Schwarz criterion 3.619033
Log likelihood -543.7404 Hannan-Quinn criter. 3.575623
F-statistic 642.5535 Durbin-Watson stat 1.981716
Prob(F-statistic) 0.000000

The time plot of monthly inflation series presented on for D(FOREX) that, Forex at first difference exhibits mean
Figure 1 does not exhibit a deterministic trend and revision. The non-stationary characteristics exhibit by
appears to wander and return to its mean. The results of Forex can be attributed to its instability over time.
Augmented Dickey Fuller (ADF) unit root tests for The task of fitting forecasting models to predict the
inflation series rejected the hypothesis of unit root are relationship between Inflation and Forex or, how Forex
rejected at it level. This confirms the stationarity of has impacted on Nigeria’s inflation in this study is based on
inflation time series as shown on the time plot [Figure 1]. two considerations. Firstly, the assumption that forex in
Forex on the other hand, exhibits systematic trend and Nigeria is not regulated and is allowed to float by the
does not return to its path as shown on Figure 2. The market forces which of course, are the current happening
ADF unit root test results on Table 1 rejects the in the economy. The second consideration is targeted at
hypothesis of unit root at its level but failed to reject after determining inflation outcomes in the economy if Forex is
the first difference. This result is also shown on Figure 3 fixed or regulated.

Publication of College of Science, Joseph Sarwuan Tarka University, Makurdi

https://fuamjpas.org.ng/ 35
Ikughur et al. FUAMJPAS 5(1):30-37 June, 2025

After entertaining several ARIMAX family model with References

Forex as exogenous variable, the SARIMAX (1,0,1)(1,0,1)12
provided the “best” estimates of parameters and forecasts [1] Ibrahim, M. (2019). Inflation.
when forex is allowed to float as is currently the case.as https://www.researchgate.net/publication/332763
shown on Table 3. In this case, the coefficient of forex is 097
0.00805 (P<0.05) showing that the impact of Forex on
inflation is statistically significant. [2] Ali, A.K. and Asfaw, D.M. (2003). Nexus
The correlogram of the residuals shows that the model between Inflation, Income Inequality and
provided a good fit as the computed values all fall within Economic Growth in Ethiopia. Plosone, 16(11)
the confidence limits. The Breusch Geoffrey Test for serial
correlation also confirm there is no serial correlation in [3] Begg,D; Vernasca, G; Fischer, S and Dornbussch,
the residuals as F=0.164 (P =0.8482). The Breusch-Pagan R. (2014). Economics, 11th Edition. McGraw Hill,
Geoffrey Test for heteroskedasticity shows that the London Pp 693.
residuals are heteroskedastic as F=13.25 and P = 0.0003.
Similarly, the assumption of normality of residuals is [4] Purnomo A. (2017). The Effect of Inflation on
rejected (JB=279.6, P = 0.000). the Currency Exchange Rate Seen in The Islamic
For the model with DForex on Tables 1 and 4, the Finance. Journal Ekonomi dan Perbankan Syariah
SARIMAX(1,0,1)(1,0,1)12, with the coefficient of 0.004 is http://muqtasid.iainsalatiga.ac.id DOI:
statistically not significant as P =0.378. The hypothesis of http://dx.doi.org/10.18326/muqtasid.v8i1.42-53
homoscedasticity is accepted while the hypothesis of the
presence of serial correlation is rejected. The result shows [5] Dornbusch, R. (1976). Expectations and
that there is a reduction of impact of Forex on inflation exchange rate Dynamics. The Journal of Political
rate by approximately 100(0.004/0.00805) or 50% if there Economy, 84(6):1161-1176.
is a policy change from allowing market forces to
determine forex to having regulated or fixed forex regime [6] Svensson, L.E.O. (2000). Open-economy ınflation
On the issue of non-stationarity of forex in Nigeria’s targeting. Journal of International Economics.
economy, [26] and [27] noted that such behavior suffers (50):155–183.
permanent effects from random shocks and have no [7] Okpara, G. C. (2006), Monetary Policy Shocks
tendency to return to a long-run deterministic path. and Econometric Analysis of Nigerian
Considering the regulated Forex market regime where Experience, Nigerian Journal of Banking and
price are fixed or stable, the reduction on its impact on Finance, 6. CECTA Nigeria Limited.
inflation by about 50% is somehow in line with the views
of [13], [17] just as the case of Turkey [19], Iran [20], [8] Ogiogio, Genevesi O. (1996). A Statistical
Indonesia [4] and Nigeria [23]; [24] and [25] just as [28] Analysis of Foreign Exchange Rate Behaviour
reported that an early warning tool based on unit root in Nigeria’s Auction” African Economic
methods provides a valuable accessory in financial stability Research Consortium, Research Paper 49.
supervision while [27] asserted that the hypothesis of
non-stationarity is robust to the inclusion of breaks. [9] CBN (2021). Understanding Monetary Policy
Series 8. www.CBN.gov.ng.
Conclusion
In this study, SARIMAX (1,0,1)(1,0,1) 12 has been found [10] Hill, C. N. L. (2005), International Business
to provide the best prediction of inflation rate in (Competing in the Global Market
Nigeria considering Forex (is allowed to float) as Place),McGraw –Hill.
exogenous variable. The findings in this study suggests
that foreign exchange (Forex) when uncontrolled, [11] Ajekwe, T., Korna, J. M., and Idyu, I.A. (2013). A
affects inflation in Nigeria as consistently with reported Comparative Analysis of the Impact of the
by [23], [24] and [25] among others while [8] suggested Fixed and Flexible Exchange Rate Systems on
preference for fixed foreign exchange (Dforex) for a the Nigerian Economy. IOSR Journal of Business
developing economy in order to combat inflation as it and Management 14(3):73-92.
has an insignificant impact on inflation in Nigeria. Thus,
there is need for intervention through monetary policy [12] Otori, E. (2023). How Forex Impacts Trade in
targeted at stabilizing Forex as its volatility, no matter Nigeria, Business Day Newspaper of January 2,
how small has impact on inflation rate in Nigeria. 2023.
This implies that controlling forex market operation of
dollars will significantly reduce rate of inflation in [13] Barro, R.J. and Gordon, D.B. (1983), “Rules,
Nigeria. Policy makers need to look at these among Discretion, and Reputation in a Model of
other macroeconomic and social variable that tend to Monetary Policy”, Journal of Monetary Economics,
affect inflation in Nigeria with the view to tuning them 12, July,101-121.
to stability.

Publication of College of Science, Joseph Sarwuan Tarka University, Makurdi

36
https://fuamjpas.org.ng/
 Ikughur et al. FUAMJPAS 5(1):30-37 June, 2025

[14] Giavazzi, Francesco and Giovannini, Alberto, Economics and Finance, 3(1), 26–38.
(1989), Limiting Exchange Rate Flexibility, https://doi.org/10.33094/26410265.2021.31.26.38
Cambridge, the MIT Press.
[25] Ighoroje, J.E and Ogheneovo Orife, C. (2022).
[15] Velasco A., (1996), “Fixed Exchange Rates: Exchange Rate Fluctuations and Inflation Rate in
Credibility, Flexibility and Multiplicity”, European Nigeria: 1987 To 2019. International Journal of
Economic Review, 40[3-5], April, 1023-1035. Business and Management Invention 11(5):01-
[16] Ghosh, A.R., Ostry, J.D., Gulde, A.M. and Wolf, 08
H.C., (1997), “Does the Nominal Exchange
[26] Nelson, C. and Plosser, C. (1982). Trends and
Rate Regime Matter?” NBER Working
Random Walks in Macroeconomic Time Series:
Paper.5874
Some Evidence and Implications, Journal of
[17] Toulaboe, D. and Terry, R. (2013), “Exchange
Monetary Economics, 10:139-169.
Rate Regime: Does it Matter for Inflation”.
Journal of Applied Business, and Economics, 14(1) [27] Libanio, G.A. (2005). Unit Roots in
http://www.wtrg.com/prices.htm Macroeconomic Time Series: Theory,
Implications and Evidence. Nova
[18] Choudhri, E. U., and Hakura, D. S. (2006). Economia_Belo Horizonte, 15(3):145-176.
Exchange Rate Pass-Through to Domestic Prices:
Does the Inflationary Environment Matter, [28] Virtanen, T; Tölö, E; Virén, M and aipalus,
Journal of International Money and Finance, 25: K.(2017). Use of unit root methods in early
614-639. warning of financial crises. European System
[19] Volkan, A; Satcioglu, C. and Korap, L.. (2007). Risk Board, Working Paper Series, No 45 / June
Impact of Exchange Rate Changes on Domestic 2017. Pp35
Inflation: The Turkish Experience. Econstor; [29] Akay, E.C. Oskonbaeva, Z and Bülbül, H. (2020).
www.tek.org.tr. DiscussionpPater2007/6. What do unit root tests tell us about
Pp10. unemployment hysteresis in transition
economies? Applied Economic Analysis
[20] Monfared, S.S and Akin, F.(2017). The 28(84/2):221-238 Emerald Publishing Limited
Relationship Between Exchange Rate and 2632-7627 DOI 10.1108/AEA-05-2020-0048
Inflation: The Case of Iran. European Journal
of Sustainable Development, 6(4):329-340 [30] Darne, O. and Diebolt, C. (2005).
NonStationary Test in Macroeconomic Time
[21] Durevall, D. and Sjo, B. (2012) 'The Dynamics of
Series.www.researchgate.net/publication/28014
Inflation in Ethiopia and Kenya'. IMF Working
9570, 11 September 2015, pp24.
Paper WP/25/66. Online
[31] Engle, F. R, Granger, C.W. (1991). Cointegration
[22] Okoli, T.M. Mbah, S.A and Agu,O.C. (2016).
and Error Correction: Representation,
Exchange Rate Volatility and Inflation: The
Estimation and Testing. Econometrica, 55 (2):
Nigerian Experience. Journal of Economics and
252–56.
Sustainable Development, 7(10)
[32] Ljung G. M. & Box G. E. P. (1978). "On a
[23] Abdulhamid, J; Musa, I and Rabiu, S. (2022).
Measure of a Lack of Fit in Time Series Models".
Inflation and Exchange Rate in Nigeria Confluence
Biometrika. 65 (2): 297–303.
Journal of Economics and Allied Sciences, 5(2):130-
141
[33] Brockwell, P. J., Davis, R A. M. & Davis, R. J.
[24] Nuhu, M. (2021). Impact of Exchange Rate
(2002). Introduction to Time Series and
Volatility on Inflation in Nigeria. Journal of
Forecasting. p. 36. ISBN 978-0-387-95351-9.
Contemporary Research in Business,

IKUGHUR J.A., UDOUMOH E.F. and ODUNEWU E.A. (2025). The Impact of Fixed and Floating Foreign Exchange
Rates on Inflation Rates in Nigeria: Statistical Evidence FUAM Journal of Pure and Applied Science, 5(1):30-37

© 2025 by the author. Licensee College of Science, Joseph SarwuanTarka

University, Makurdi. This article is an open access article distributed under the terms
and conditions of the Creative Commons Attribution (CC) license.

37

View publication stats