









































Pa
ge

 
1



Pa
ge

 
24

American Journal of  Applied 
Statistics and Economics (AJASE)

Interrogating the Asymmetric Impact of  Exchange Rate on Agricultural Output in Nigeria
Francis Ariayefa Eniekezimene1*, Ebimowei Wodu2, Mr Joseph Peres Anda-Owei3

Volume 3 Issue 1, Year 2024
ISSN: 2992-927X (Online)

DOI: https://doi.org/10.54536/ajase.v2i1.2247
https://journals.e-palli.com/home/index.php/ajase

Article Information ABSTRACT

Received: November 26, 2023

Accepted: December 29, 2023

Published: December 31, 2023

The agricultural sector is vital to the Nigerian economy. It serves several functions including 
food supply for the populace, a source of  income for a sizable section of  the domestic labour 
force, a supplier of  raw materials for industrial production, and provision of  foreign exchange 
revenues to the economy. However, due to the country’s abundance of  natural resources 
like crude oil, the agriculture sector is vulnerable to the fluctuations in the exchange rate 
caused by international financial markets. This analysis looked at how changes in the value 
of  the naira relative to dollar affects agricultural production. In order to contribute to the 
current body of  literature for the time period between 1981 and 2021, this study employed 
the Dutch Disease Syndrome theoretical framework through the use of  the non-linear 
autoregressive distributed lag (NARDL) method for Nigeria. This research shows that while 
the long-term effect of  the exchange rate on agricultural output in Nigeria is symmetrical, 
the short-term effect is asymmetrical. However, in more specific terms, the symmetrical 
effect of  the exchange rate in the long run revealed that exchange rate appreciation 
increased real agricultural GDP by approximately 8.8 percent compared to exchange rate 
depreciation which had only increased real agricultural GDP by 0.11 percent. Consequently, 
since exchange rate exerted positive impact on agricultural production in the long run, it 
is suggested that the Nigerian government explores the increased competitiveness of  the 
agricultural sector in its economic diversification efforts. In other words, the agricultural 
sector could provide an avenue to expand the revenue base of  the government, but for a 
more beneficial effect on the agricultural sector in particular and the economy in general, 
more focus should be placed on policies that would enhance appreciation of  the Naira such 
as reducing imports of  agricultural inputs and produce.

Keywords

Exchange Rate, Asymmetric 
Impact, Agricultural Output, 
Dutch Disease Syndrome, 
Non-Linear Autoregressive 
Distributed Lag

1 Department of  Economics, Niger Delta University, Wilberforce Island, Bayelsa State, Nigeria
2 Bayelsa State Ministry of  Finance, Yenagoa, Bayelsa State, Nigeria
3 Department of  Economics, Nile University of  Nigeria, Abuja
* Corresponding author’s e-mail: franciseniekezimene@ndu.edu.ng

INTRODUCTION
Agriculture encompasses not only the raising of  livestock 
and aquatic organisms for human consumption and 
industrial production, but also the exploration and use of  
forests for these same ends. Since agriculture accounts 
for a significant portion of  Nigeria’s GDP and provides 
millions of  people with jobs, it is vital to the country’s 
economy. To be more specific, agriculture helps the 
industrial sector by providing surplus labour, food for 
domestic consumption, a market for industrial output, 
domestic savings for industrial investment, and foreign 
exchange earnings from agricultural exports. Based on the 
latter argument, it’s reasonable to assume that changes in 
the value of  the naira can have effect on the profitability 
and competitiveness of  agricultural output in Nigeria.
Conversely, the appreciation or depreciation of  one 
nation’s currency relative to the economies of  other 
countries has an effect on the agricultural production 
and balance of  payment of  the country providing the 
data. Economic factors and sectors including export 
profits, the cost of  imported inputs, food price inflation, 
investment and production choices, and government 
policies are all susceptible to the effects of  swings in 
exchange rates. Nigeria’s agricultural exports may become 
less competitive as a result of  fluctuations in the exchange 
rate. When the naira weakens against major international 

currencies like the US dollar, Nigerian agricultural 
exports become more competitive on worldwide markets 
(Akinbode & Ojo, 2018). Export earnings for Nigeria’s 
farmers may improve if  foreign buyers can get more of  
the country’s food for the same amount of  money in their 
currency.
Because of  its reliance on foreign inputs like herbicides, 
equipment, and fertilisers, Nigeria’s agricultural business 
can also be hit by fluctuations in the price of  these 
items. Input prices may be affected by a depreciation of  
the local currency due to the higher price of  imported 
goods (Onyeagocha el al., 2021). Farmers’ profitability 
and competitiveness may suffer as a result of  higher 
production costs. Changes in the value of  the naira might 
potentially have an effect on the cost of  food in Nigeria. 
When the local currency gains, the cost of  imported 
goods rises (Onwuka & Oyewumi, 2019). Consumers 
may be harmed, especially those with lower means who 
may have a harder time locating reasonably priced food 
that is also healthy. The uncertainty caused by changes in 
the value of  a currency might affect agricultural output 
and investment decisions. It’s possible that farmers and 
agribusinesses won’t invest in long-term projects or boost 
output if  the currency rate remains volatile (Onyeagocha et 
al., 2021). Reduced agricultural production and slow sector 
growth are possible outcomes. Currency fluctuations may 



Pa
ge

 
25

https://journals.e-palli.com/home/index.php/ajase

Am. J. Appl. Stat. Econ. 3(1) 24-32, 2024

have a significant impact on the agriculture industry, thus 
governments sometimes step in to stabilise the market 
or mitigate the fallout. The Central Bank of  Nigeria, 
for instance, may employ capital controls or currency 
interventions to dampen fluctuations in the value of  the 
naira (Onwuka & Oyewumi, 2019). These bills are an 
attempt to guarantee the agriculture sector’s continued 
success and expansion over the long term by providing 
stability and aiding it.

Stylized Facts about Nigeria’s Agricultural Sector 
The government has used both monetary and fiscal 
measures to help agriculture regain its former glory and 
prevent its role in the economy from shrinking further. 
The Nigerian Agricultural Bank (NAB) was founded in 
1972 as a direct result of  government policies and projects. 
Before becoming the Bank of  Agriculture (BOA) in 2000, 
it was known as the Nigerian Agricultural and Cooperative 
Bank (NACB) from 1978 to 1978. It wasn’t until 1976 that 
initiatives like the Agricultural Development Programme, 
the River Basin Development Authority, and Operation 
Feed the Nation came into being. Although the World 
Bank Agricultural Development Projects (ADP) were 
established before 1976, they were originally conceived as 
an all-encompassing programme for rural development. 
The Agricultural Credit Guarantee Scheme (1977), Rural 
Banking Programme (1977), Green Revolution (1979), 
Directorate for Food, Roads, and Rural Infrastructure 
(1986), National Fadama Development Project (1990), 
National Special Programme on Food Security (NSPFS) 
(2000), Community-Based Agricultural and Rural 
Development Schemes (2003-2008), and Root and Tuber 
Expansion Programme (RTEP) (2004) are all examples 
of  government initiatives in the agricultural sector 
(Eniekezimene, Oluwabusayo & Ekiye, 2020).
Although the government has taken steps to improve 
agriculture, the sector’s performance over the years has 
been dismal. Agriculture’s contribution to GDP is lower 
than that of  industry and services, and there is a clear 
divergence between agriculture and the manufacturing 
sectors. Foreign reserves are depleted and the exchange 
rate is weakened because of  the country’s reliance on 
imports of  agricultural items such as rice, frozen chicken, 
iced fish, and plywood, to name a few. The government 
has banned over 40 categories of  imports, most notably 
food products, and these items have since been removed 
off  the list of  recognised commodities that may be 
purchased with foreign exchange, but the situation has not 
improved. Additionally, considering the need to relinquish 
complete reliance on crude oil and to emphasize on the 
non-oil sector in driving economic growth in Nigeria, 
it has become imperative to re-examine the effect of  
exchange rate asymmetry on productive sectors like the 
agriculture in Nigeria. As a result, the purpose of  this 
research is to examine the impact of  fluctuating exchange 
rates on agricultural output in Nigeria between the years 
1981 and 2021. The major objective of  this research is to 
analyse the impact that the naira’s exchange rate during 

the study period had on agricultural output in Nigeria. 
The specific objectives of  this paper include to examine 
the impact of  average manufacturing capacity utilization 
rate on agricultural output in Nigeria and to study the 
impacts of  inflation and interest rates on agricultural 
production in Nigeria. 
The sample period used in this study was found to follow a 
normal distribution when degrees of  freedom were taken 
into account. Therefore, a 41-year period, beginning in 
1981 and ending in 2021, was selected for analysis. From 
an empirical perspective, this time frame encompasses the 
vast majority of  exchange rate fluctuations experienced 
by Nigeria after it gained independence.  The following 
are components of  the research’s later stages: The second 
section reviews the relevant literature and theory, while 
the third discusses methodology in detail. In section 
four, we provide a detailed review of  the data and the 
findings. Section five concludes with some last thoughts 
and suggestions for follow-up study.

Theoretical Framework and Empirical Literature 
Review
The Dutch Disease Syndrome model created by Corden 
in 1984 was employed in this study, following the 
example of  Adekunle, Kehinde, & Taiwo (2019). The 
model assumes three sectors: non-tradable (N), which 
is analogous to Nigeria’s service industry; lagging (L), 
which is analogous to Nigeria’s agriculture sector; and 
booming (B), which is analogous to Nigeria’s oil and gas 
firm. Furthermore, the model assumes that sectors B and 
L exist, as they are the ones responsible for producing 
commodities that are traded at standard world prices. 
Third, a multiskilled workforce that can work across all 
three industries to keep costs down while yet contributing 
to each one’s production and unique characteristics.
In addition, factor price is not static worldwide, which 
brings us to our fourth and last reason. Based on these 
projections, the model creates two distinct consequences 
of  a boom in B: the spending impact and the resource 
migration effect. As a result, the starting wages of  the 
first generation of  factor employees have increased. The 
price of  non-traded products (N) will increase relative 
to the price of  traded goods if  some of  the additional 
money created by B is spent, either by factor owners or by 
the government in the form of  taxes. This is an example 
of  sincere thanks. To achieve this result, one must 
reallocate resources from B and L into N and reverse 
the flow of  demand from N to B and L. Consequences 
of  redistributing wealth: Since the marginal product of  
labour in B has gone up because of  the boom, workers in 
L and N are leaving in pursuit of  better job security and 
a more consistent income in terms of  traded products. 
This finding may be broken down into two parts: When 
employees from L are moved to B, production in L falls. 
This might be seen as direct de-industrialization because 
there is no demand for a real appreciation of  the currency 
rate and no market for N. Workers in N may easily migrate 
to B since the real exchange rate is stable. The transfer 



Pa
ge

 
26

https://journals.e-palli.com/home/index.php/ajase

Am. J. Appl. Stat. Econ. 3(1) 24-32, 2024

of  resources boosts real appreciation, and the resulting 
spending impact raises N’s excess demand. As a result, the 
deindustrialization caused by the cost impact intensifies as 
more people move from L to N in search of  work. Indirect 
deindustrialization describes this secondary effect.

Empirical Literature Review
While this may be true, the relationship between the 
exchange rate and agricultural output has been the subject 
of  several empirical studies, both globally and domestically. 
In order to stay abreast with the empirical results and 
advancements, this section evaluated a few of  these 
investigations, beginning with those conducted abroad and 
concluding with those conducted in the United States. For 
instance, Reuben and Alala (2014) used time series data 
from 1970 to 2008 to analyse the effect of  currency rate 
fluctuation on the prosperity of  Kenyan tea exports. The 
ADF and Johansen Cointegration were used to guarantee 
series stationarity and the long run relationship of  the 
variable, and then the ECM approach was used to examine 
the empirical model. Results demonstrated that exchange 
rate volatility harmed tea exports. Sirikul, Chanchai, and 
Somchai (2015) analysed the effect of  monthly exchange 
rate variations on Thai rice and rubber exports using 
primary and secondary data from 2002 to 2014. Using a 
combination of  survey data and in-depth interviews, they 
determined that fluctuations in the value of  the currency 
exchange rate dampened demand for both types of  
agricultural exports. The study found that exporters and 
firms might benefit from hedging their risk of  currency 
rate fluctuations. 
Kafle and Kennedy (2015) investigated how the real 
exchange rate affects agricultural exports from the 
United States to the nations of  the Organisation for 
Economic Cooperation and Development (OECD). The 
consequences of  free trade agreements and embracing the 
Euro as a national currency were also studied. From 1970 
to 2010, bilateral trade flow panel data were analysed using 
the Gravity Model. Both agricultural and non-agricultural 
trade flows were demonstrated to be negatively affected 
by fluctuations in the real exchange rate. The export-
dependent non-agricultural businesses are particularly 
vulnerable to variations in the exchange rate.
In their 2018 study, Wagan, Chen, Seelro, and Shah looked 
into how changes in monetary policy affected expansion, 
inflation, and job creation. Using a factor-augmented 
vector autoregressive model established by Bernanke 
et al. (2005), researchers analysed agricultural data for 
Pakistan and India from 1995 to 2016 and found that 
restrictive monetary policy dramatically increased rural 
unemployment while significantly reduced food inflation 
and agricultural productivity. Mashinini, Dlamini, and 
Dlamini (2019) studied the effects of  monetary policy 
on agricultural output in Eswatini from 1980 to 2016. 
Researchers employed a technique known as the vector 
error correction model (VECM). Statistics show that 
currency rate, interest rate, inflation rate, money supply, 
and agricultural loans all have a detrimental effect on 

agricultural output. However, in the near run, agricultural 
output benefited from these causes.
Kipruto and Nzai (2018) studied the results of  government 
investment on agricultural output in Kenya from 1980 to 
2016. When measured with the autoregressive distributed 
lag (ARDL) technique, government spending increases 
agricultural production.
Gatawa and Mahmud (2017) looked at the long- and short-
term effects of  currency rate changes on the volume of  
Nigeria’s agricultural exports between 1981 and 2014. 
The study used estimation strategies like generalised 
autoregressive conditional heteroskedasticity (GARCH) 
and autoregressive distributed lag (ARDL) to determine 
the relationship between the exchange rate, agricultural 
loan amounts, and agricultural export relative prices and 
the volume of  agricultural exports. Long-term agricultural 
output was shown to be favourably influenced by the 
exchange rate and agricultural loans, whereas short-term 
agricultural output was negatively impacted by the relative 
pricing of  export agricultural commodities. Long-term 
results were similar, with the exception of  agricultural 
exports, which were significantly impacted negatively by 
the official exchange rate. 
Akinbode and Ojo (2018) utilised GARCH and ARDL to 
examine the relationship between the exchange rate and 
Nigeria’s agricultural exports from 1980 to 2015. Inflation, 
GDP growth, inflation, and global pricing were shown to 
have a significant positive influence on agricultural exports, 
whereas the short- and long-term effects of  exchange 
rate fluctuations on agricultural production were found 
to be minimal. The dynamic impact of  exchange rate 
fluctuations on agricultural output in Nigeria was studied 
by Adekunle, et al (2019) using the non-linear autoregressive 
distributed lag (NARDL) approach between 1981 and 
2018. Although the bounds test indicated that there was 
no long-term relationship between the dependent and 
set of  independent variables, their findings indicated 
that real exchange rate appreciation and depreciation, as 
well as the set of  explanatory variables, had a significant 
impact on agricultural output in the short run. Alegwu, 
Aye, and Asogwa (2018) examined the correlation 
between the fluctuation of  the Nigerian currency and 
agricultural exports from 1970 to 2013 using the vector 
error correction model (VECM). The data showed that 
fluctuations in currency exchange rates had a negative 
impact on agricultural exports over the long run but had 
no impact over the short term.  
Ochalibe, and Enete (2019) used the Granger Causality 
approach to analyse the impact of  the currency rate and 
the interest rate on agricultural output growth in Nigeria 
from 1980 to 2018. The findings indicated a unidirectional 
relationship between the interest rate and the exchange 
rate and agricultural progress. The impact of  the exchange 
rate policy tool was positive by 2.85%, but interest rates 
considerably hampered agricultural development.
Ikpesu and Okpe (2019) examined the relationship between 
capital flows, currency rates, and agricultural output in 
Nigeria using the autoregressive distributed lag (ARDL) 



Pa
ge

 
27

https://journals.e-palli.com/home/index.php/ajase

Am. J. Appl. Stat. Econ. 3(1) 24-32, 2024

method between 1981 and 2016. A decline in currency 
value, the study revealed, has both short-term and long-
term negative effects on agricultural output. Abubakar 
(2019) looked explored the correlation between Nigeria’s 
interest rates and agricultural output between 1999 and 
2016. The numbers showed that when interest rates rose, 
agricultural output fell. 
Awolaja and Okedina (2020) examined how changes in 
exchange rates affected agricultural output in Nigeria. 
In this study, the authors employed non-linear auto-
regressive distributed lags (NARDL) technique. Long-
term agricultural output was shown to be greatly increased 
by a real exchange rate appreciation, whereas agricultural 
output was significantly decreased by a real exchange rate 
depreciation. Another finding from the forecasts was that 
agricultural output is more vulnerable to increases in the 
real exchange rate than decreases in the same variable. 
This study, in line with the work of  (Adekunle et. al, 2019), 
used exchange rate appreciation and depreciation alongside 
interest rate, inflation rate, and average manufacturing 
capacity utilization rate to examine the asymmetric 
influence of  exchange rate on agricultural production in 
Nigeria. However, the current study focused on monetary 
policy variables (interest rate and inflation) and agricultural 
FDI to determine the impact on agricultural output, in 
contrast to the work of  (Adekunle et al, 2019) who used 
exchange rate appreciation and depreciation alongside 
fiscal and monetary policies as well as the industrial sector 
capacity utilization rate as explanatory variables.
   
Methodology and Model Specification
This study employed the non-linear autoregressive 
distributed lag (NARDL) method. The time-varying 
impact of  changes in independent variables on the 
dependent variable is more accurately captured by the 
nonlinear autoregressive distributed lag (NARDL) than by 
the more frequent autoregressive distributed lag (ARDL). 
Time series data for the variables were gathered from the 
Central Bank of  Nigeria (CBN) Statistical Bulletin from 
1981 to 2021, and stationarity was tested using the ADF 
unit root test, and long run correlations were determined 
using the Bounds test for cointegration. Therefore, the 
operational model for this study is:   
ARGDP =f(NER, INR, INFR, MCU)                      (3.1)
The econometric form of  equation (3.1) which considers 
the dynamics of  exchange rate is as follows:
LARGDPt= β0+β+

1NER+
t+β-

1NER-
1+β2INRt+β3INFRt 

+β4 MCUt+µt                                     (3.2)
The asymmetric impact of  exchange rate on agricultural 
output using NARDL and by following the approach of  
Shin, Yu and Greenwood-Nimmo (2014), yields; 
∆ARGDP t=γ[ARGDP t+β+

1NER+
(t-1)+β -

1NER-
(t-1)+ 

β2INR(t-1)+β3INFR(t-1)+β4MCU(t-1)+∑(q1-1)
(j=0)δ

+
j∆NER+

(t-j) 

+∑ (q2-1
)(j=0)δ

-
j∆NER -

(t-j)+∑ (q3-1)
(j=0)λ jINR+∑ (q4-1)

(j=0)π j 

∆INFR(t-j)+∑(q5-1)
(j=0)ϕj ∆MCU(t-j)+µt                         (3.3)                                                                                                               

Equation (3.3) can be re-parameterized to derive the 
unrestricted error correction version below;

∆ARGDPt = γ[ARGDP(t-1)-(-β1/γ NER+
(t-1) - β1/γ NER-

(t-1)- β2/γ INR(t-1)- β3/γ INFR(t-1)-β4/γ MCU(t-1))+∑ (p-1)

(i=1) θi ∆ARGDP(t-i )+∑(q1-1)
(j=0) δ

+
j ∆NER+

(t-j)+∑ (q2-1)
(j=0)δ

-
j 

∆NER-
(t-j)+∑(q3-1)

(j=0)λj INR+ ∑(q4-1)
(j=0)πj ∆INFR(t-j)+ 

∑(q5-1)
(j=0)ϕj ∆MCU(t-j)+μt                                                   (3.4)                                                                                                   

By letting, 
E(t-1)=ARGDP(t-1)-β

+
1NER+

(t-1)-β
-
1NER-

(t-1)-β2INR(t-1)- 
β3INFR(t-1) - β4MCU(t-1)                   (3.5)
Where,
β+

1= -α1/γ, β-
1 = -α1/γ, β2= -α2/γ, β3= -α3/γ, β4= -α4/ γ        (3.6)

Equation (3.4) then, becomes,
∆LARGDPt= γE(t-1)+∑(p-1)

(i=1)θi ∆ARGDP(t-i)+∑(q1-1)
(j=0)δ

+
j 

∆NER+
(t-j)+∑(q2-1)

(j=0)δ
-
j∆NER-

(t-j)+∑(q3-1)
(j=0)λjINR(t-j)+∑(q4-1)

(j=0)πj ∆INFR(t-j)+∑(q5-1)
(j=0)ϕj ∆MCU(t-j)+μt                                (3.7)

Where:
Δ = first difference operator; 
t = time period; 
LARGDP = Logged share of  agriculture in real GDP; 
E(t-1) = error correction term 
N𝐸𝑅 = nominal exchange rate; NER+ = Positive changes 
in nominal exchange rate (representing depreciation); 
NER- = Negative changes in nominal exchange rate 
(representing appreciation);
INR = Interest rate (a proxy for the role of  monetary 
policy in the agricultural sector development) 
INFR = Inflation rate (as a control variable)
MCU = Average manufacturing capacity utilization % (to 
account for intersectoral linkage)
θi, δj+, δj-, 𝝀j, πj, ϕj, are short run parameters, while 
β1,………, β4 are long run parameters.
p is the lag length for the dependent variable, while 
q1,…… ,q4 are the lag lengths associated with the 
explanatory variables, and μ = random error term.

A Priori Expectations/Expected Results 
The preposition of  economic theory regarding the 
relationship between each of  the characteristics of  the 
explanatory factors and the dependent variable serves as 
the foundation for the a priori expectations. The short 
run and long run relations of  these parameter coefficients 
are respectively shown as:
δ+

j > 0 or < 0, δ-
j > 0 or < 0, 𝝀j < 0, πj > 0, ϕj > 0,

β+
1 > 0 or < 0, β-

1 > 0 or < 0, β2 < 0, β3 > 0, β4 > 0,
Theoretical speculation suggests that an increase or 
reduction in agricultural output may result from an 
appreciating currency. Devaluation or appreciation of  a 
currency’s exchange rate has similar effects on agricultural 
production. As the cost of  borrowing money rises, 
it is anticipated that agricultural output would fall as a 
result of  an increase in interest rates. Higher prices for 
agricultural produce are anticipated to boost production 
and supply, therefore increasing the inflation rate. 
Similarly, an increase in average manufacturing capacity 
utilization is expected to increase agricultural productivity 
because of  the link between the agricultural sector and 
the manufacturing sector. 



Pa
ge

 
28

https://journals.e-palli.com/home/index.php/ajase

Am. J. Appl. Stat. Econ. 3(1) 24-32, 2024

Empirical Analysis
Descriptive Statistics
The descriptive statistics for the five variables in this 
study are summarised in Table 1, which spans the years 
1981 through 2021 and represents 41 years of  data. The 
average value of  the exchange rate is close to 116.52%, 

whereas the average value of  the log of  agricultural real 
GDP is close to 8.8%. Standard deviations of  the series 
from their respective means show that the exchange 
rate was the most volatile (approximately 108.78%), and 
the log of  real agricultural GDP was the least volatile 
(approximately 0.73%).

Table 1: Summary of  Descriptive Statistics
Variable No of  Obs. Mean Maximum Minimum Std. Dev
LARGDP 41 8.7989 9.8383 7.7422 0.7271
NER 41 116.5188 403.0000 0.6100 108.7793
INR 41 17.2517 29.8000 7.7500 4.6270
INFR 41 18.9334 72.8400 5.3800 16.6654
MCU 41 11.7708 14.1637 8.2316 1.9702

Source: Author’s computation

Unit Root Test Result
The assumption behind the unit root test is, of  course, 
that a series has a unit root. Table 2 shows that the 
only variable that passed the Augmented Dickey Fuller 
(ADF) unit root test and remained stable at levels was 

the inflation rate (INFR). When we differentiated the 
dependent variable, the other four variables (agricultural 
real GDP, average manufacturing capacity utilization, the 
nominal exchange rate, and the interest rate) remained 
unchanged.

Table 2: ADF Unit Root Test Result
Variable ADF Statistics Probability I(d)

5% Critical val Levels First Difference Levels First Difference
ARGDP -3.526609 -1.918614 -5.619461***  0.6263  0.0002 I(1)
NER -3.526609 -0.368921 -5.285607***  0.9855  0.0006 I(1)
INR -3.526609 -0.368921 -5.285607***  0.9855  0.0006 I(1)
INFR -3.529758 -4.097623*** Ψ…….  0.0133 Ψ……. I(0)
MCU -3.529758 -1.608017 -3.568379***  0.7715  0.0082 I(1)

Please take notice that ***, **, and * indicate that the unit root hypothesis was rejected at the 1%, 5%, and 10% levels, respectively. 
The number of  iterations via differentiation that must occur until a series becomes stationary is denoted by the order of  integration, I(d). 
Only intercept models are considered.
Source: Author’s computation

The ARDL Bounds Test for Cointegration Results
After learning about the time series, we validated the long-
term connection. The null hypothesis of  no cointegration 
between the variables underlies the Bounds test for 
cointegration. Using the Autoregressive Distributed Lag 
(ARDL) model and the Bounds test, we checked for 

long-term correlations between the series. Because it 
influences the outcome of  the ARDL processes, the lag 
time was selected with care. 
This research followed the recommendation of  Pesaran 
et al. (2001) and used AIC to establish the time lag. 
Therefore, the selected ARDL model (2,3,4,4,3) was 

Table 3: ARDL Bounds Test for Cointegration Result
F-statistic = 5.607080; No. of  Parameters K = 4

Critical Bounds Values
Level of  Significance Lower Bounds I(0) Upper Bounds I(1)
10% 2.2 3.09
5% 2.56 3.49
2.5% 2.88 3.87
1% 3.29 4.37

Source: Author’s computation 

used to investigate the long-term relationship between 
each variable. Non-linear findings from the Bounds test 

for cointegration between agricultural production and 
its possible drivers (positive and negative changes in 



Pa
ge

 
29

https://journals.e-palli.com/home/index.php/ajase

Am. J. Appl. Stat. Econ. 3(1) 24-32, 2024

exchange rate, FDI, interest rate, and inflation rate) are 
shown in Table 3. The long-term relationship between 
agricultural production and its causes is supported by 
the fact that the F-statistic of  5.607080 is larger than 
the upper I(1) critical bound of  4.37 at the 1% level of  
significance.

Estimation and Discussion of  Results 
Long Run Estimate of  the NARDL Model
After confirming the existence of  a long-term relationship 
between the variables, we proceeded to estimate the long-
run coefficient estimates in Equation (3.2). We evaluated 
long-run elasticities using the AIC. Table 4 displays the 
results of  NARDL’s long-term estimations of  the model’s 
parameters. There was a dissection of  the exchange rate’s 
asymmetry. Table 4 shows that at the 5% level (0.0354), 
the positive coefficient of  0.001100 for the exchange rate 
denoting depreciation or devaluation of  the Naira over 
the U.S. dollar is statistically significant. This indicates that 
real agricultural GDP benefited greatly from the decline 
of  the currency over time. This finding accords with what 
one would expect from a purely economic standpoint. 

However, the exchange rate was judged to be statistically 
significant at the 5% level (-0.087991), indicating that the 
Naira had appreciated against the US dollar. This indicates 
that a higher exchange rate was the long-term driver of  
a higher RAGDP. The precise figure is an 8.8 percent 
increase in real agricultural GDP for every unit decline 
in the exchange rate. At the 5% significance level, the 
interest rate coefficient was positive (0.391737), but the 
probability showed no significant relationship (0.6104). 
There was no statistically significant relationship between 
the inflation rate and its negative coefficient (-.092174) at 
the 5% level (0.5772). However, the result conforms to 
what was expected going in, and it shows that inflation 
decreased agricultural GDP by 9.2 percent per unit 
increase in inflation over the long run. Manufacturing 
capacity utilization was positive (0.013067) and statistically 
significant at the 5 percent level of  significance (0.0351) 
based on the coefficient and probability value respectively. 
This implies that a unit increase in manufacturing capacity 
utilization increased real agricultural GDP by approximately 
1.3 percent in the long run.

Table 4: Long Run Regression Result
Dependent Variable: LARGDP
Variable Coefficient Std. Error t-Statistics Probability
NER_POS 0.001100 2.67E-05 41.21086 0.0154
NER_NEG -0.087991 0.003326 -26.45765 0.0241
INR 0.391737 0.753849 0.519649 0.6104
INFR -0.092174 0.161960 -0.569120 0.5772
MCU 0.013067 0.000722 18.10486 0.0351
C 14.64121 17.53933 0.834765 0.4161

Source: Author’s computation

The Error Correction Regression Model
Short-term and long-term projections for the ARDL and 
ECM are displayed in Table 5, respectively. The ECM 
is represented by the notation CointEq(-1). The ECM 
model examines the rate of  adjustment in response to 
deviations from the long-run equilibrium and aims to 
capture the short-term dynamics of  exchange rate and 
agricultural production. When about 43% of  the long-run 
disequilibrium is rectified by lag-period error shocks, the 
coefficient of  the error correction component is negative 
and statistically significant. Adjusted R2 = 0.965652 
indicates that about 97% of  the variance in ARGDP was 
well captured by the model’s explanatory factors.
The inclusion of  a negative autoregressive coefficient, or 
lag coefficients, for the agricultural output share indicated 
that the computation of  the agricultural element of  real 
GDP was not adaptive. At the 5% level of  significance, all 
of  the coefficients are significant. Real agricultural output 
has a major negative influence on itself  in the short to 
medium term. Exchange rate depreciation was negative 
in the reporting year and the first, second, and third 
lags (-0.000852, -0.001620, -0.003691, and -0.000101, 
respectively), with the exception of  the third lag, which 

was significant at the 10% level.
The results showed that a drop in the value of  the Naira 
compared to the US Dollar or exchange rate depreciation 
had a significant negative impact on real agricultural 
GDP in the short run. During the reporting period, the 
first lag was 0.108603, the second lag was -0.046579, the 
third lag was -0.076781, and the fourth lag was -0.084302 
with probabilities (P=0.0044, P=0.0082, P=0.0042, 
and P=0.0049, respectively), in contrast to the negative 
exchange rate. This result suggests that short-term 
negative exchange rate coefficients had a larger beneficial 
influence on real agricultural GDP in the reporting year 
than long-term positive exchange rate coefficients.
Table 5 displays the negative impact on agricultural real 
GDP in the first, second, and third lag years due to a 
shift in interest rates. According to these numbers, real 
GDP or output in agriculture dropped by nearly 4% for 
every unit rise in interest rates in the first and second 
lag years, and by about 2% in the third lag year. Table 
5 shows a similar trend, showing that higher inflation 
rates had a moderately negative effect on agricultural real 
GDP in the current year, but a substantially more severe 
effect in the first, second, and third lagged years. This 



Pa
ge

 
30

https://journals.e-palli.com/home/index.php/ajase

Am. J. Appl. Stat. Econ. 3(1) 24-32, 2024

finding suggests that for every one-unit rise in inflation, 
agricultural real GDP or output declined by around 0.6% 
in the first and second lagged years and by about 0.2% in 
the third lagged year. 
Manufacturing capacity utilization was found negative 
in the reporting year as well as the first and second lags 
(-0.008120, -0.010035, and -0.020543) respectively and 
were all statistically significant at the 5 percent levels of  
significance (0.0098, 0.0127 and 0.0057) respectively. This 
result connotes that manufacturing capacity utilization 
had significant negative impact on agricultural output in 
the short run. This also means that, there is negative inter-
sectoral linkage between manufacturing and agricultural 
sectors in Nigeria in the short run. This outcome could 
be attributed to other factors not covered in this study, 

such as poor electricity supply and other government 
policies bedevilling the manufacturing sector in Nigeria. 
With respect to the post estimation results, since all of  
these post-estimation outcomes had probabilities over 
the 0.05 criterion, we may conclude that the model did 
not display non-serial correlation in the residuals, non-
normality of  the residuals, or non-constant residual 
variance. This means that the asymmetric effect model is 
sufficient for making policy suggestions.
In figure 4.1 the Jarque-Bera result of  0.306160 with 
a probability value of  0.858061 > 0.05 requires the 
retention of  the null hypothesis of  normality of  the 
residuals. Thus, the parameter estimates of  NARDL 
model used in this study are suitable for forecasting. 

Table 5: Short Run Error Correction Regression Result
Dependent Variable: LRAGDP
Variable Coefficient Std Error t-Statistics Probability
D(LARGDP(-1)) -0.358112 0.149040 -2.402788 0.0288
D(NER_POS) -0.000852 1.57E-05 -54.37188 0.0117
D(NER_POS(-1)) -0.001620 2.62E-05 -61.85776 0.0103
D(NER_POS(-2)) -0.003691 2.89E-05 -127.5422 0.0050
D(NER_POS(-3)) -0.000101 1.69E-05 -5.951011 0.1060
D(NER_NEG) 0.108603 0.000746 145.6576 0.0044
D(NER_NEG(-1)) -0.046579 0.000601 -77.50932 0.0082
D(NER_NEG(-2)) -0.076781 0.000508 -151.1548 0.0042
D(NER_NEG(-3)) -0.084302 0.000651 -129.4751 0.0049
D(INR) -0.000455 0.004876 -0.093402 0.9267
D(INR(-1)) -0.038183 0.007329 -5.210208 0.0001
D(INR(-2)) -0.039572 0.007248 -5.460035 0.0001
D(INR(-3)) -0.024967 0.005602 -4.457091 0.0004
D(INFR) -0.000238 0.000861 -0.276746 0.7855
D(INFR(-1)) -0.005562 0.001482 -3.753005 0.0017
D(INFR(-2)) -0.006240 0.001557 -4.006883 0.0010
D(INFR(-3)) -0.001935 0.000948 -2.040804 0.0581
D(MCU) -0.008120 0.000124 -65.23956 0.0098
D(MCU(-1)) -0.010035 0.000199 -50.30550 0.0127
D(MCU(-2)) -0.020543 0.000185 -111.2717 0.0057
CointEq(-1)* -0.428397 0.015525 -16.4216 0.0041
Adjusted R2 
F-statistic
Breusch-Godfrey Serial
Correlation LM test
Breusch-Pagan-Godfrey 
Heteroskedasticity test
Wald test for asymmetry 

0.965652
191.9985[0.000000]

0.2624 [0.4000]

0.1142 [0.0824]
0.9739 [0.3409]

Source: Author’s computation

CONCLUSION
This study was set out to evaluate the asymmetric impact 
of  exchange rate on agricultural output in Nigeria from 
1981 to 2021. The Dutch Disease Syndrome (DDS) was 

used as the theoretical framework of  the study while the 
non-linear autoregressive distributed lag (NARDL) was 
adopted as the analytical technique for the agricultural 
output model. The empirical results of  the model were 



Pa
ge

 
31

https://journals.e-palli.com/home/index.php/ajase

Am. J. Appl. Stat. Econ. 3(1) 24-32, 2024

Figure 1: Jarque-Bera Test for Residual Normality

separated into two. The long run results and the short 
run results. The long run results presented in table 4 
revealed that exchange rate depreciation denoted by 
positive exchange rate (NER_POS) was positively signed 
and statistically significant implying that depreciation of  
the Naira over the US Dollar increased real agricultural 
GDP by 0.11 percent over the period of  the study. 
On the other hand, exchange appreciation denoted by 
(NER_NEG) was negatively signed and also statistically 
significant showing that appreciation of  the Naira over 
the US Dollar also impacted positively on real agricultural 
GDP. Specifically, exchange rate appreciation increased 
real agricultural GDP by approximately 8.8 percent over 
the study period. Manufacturing capacity utilization was 
positive and statistically significant, implying that a unit 
increase in manufacturing capacity utilization increased 
real agricultural GDP by approximately 1.3 percent in the 
long run. Interest rate was found positive but statistically 
insignificant contrary to a priori expectation, while 
inflation rate was negative and statistically insignificant 
but in line with a priori expectation. The short run results 
presented in table 5 on the other hand revealed that 
exchange rate depreciation dented by (NER_POS) had 
significant negative impact on real agricultural GDP for 
reporting year and all the three lag years, while exchange 
rate appreciation denoted by (NER_NEG) exhibited 
significant positive impact on real agricultural GDP for the 
three lag years. Interest rate and inflation rate were both 
negatively signed and statistically significant conforming 
with a priori expectation. However, manufacturing 
capacity utilization was contrary to a priori expectation 
as it was also negatively signed and statistically significant. 
Based on the empirically findings of  this study, we 
conclude that while the long-term effect of  the exchange 
rate on agricultural output in Nigeria is symmetrical, 
the short-term effect is asymmetrical. However, the 
symmetrical effect of  the exchange rate in the long 
run revealed that exchange rate appreciation had more 
positive and significant effect on real agricultural GDP 
with 8.8 percent compared to exchange rate depreciation 

which had only 0.11 percent significant positive effect on 
real agricultural GDP.

RECOMMENDATIONS
following the findings of  this study, the following 
recommendations for policy options could be pertinent: 

(i) It is suggested that, as part of  its efforts to diversify 
the economy, the Nigerian government investigates the 
sector’s increased competitiveness. Exchange rates have a 
positive long-term influence on agricultural productivity. 
In other words, the agriculture business may provide 
a method for the government to diversify its revenue 
sources. However, focus should be placed more on 
policies that would help the Naira appreciate such as 
reducing imports of  agricultural inputs and produce.

(ii) Local raw material procurement should be prioritised 
in order to absorb the beneficial spill over inherent in 
forward and backward inter-sectoral connections; doing 
so will cut imported inflation, total inflation in Nigeria, 
and the strain on foreign exchange. 

(iii) Genuine farmers should be eligible for low-interest 
loans and inputs from the Nigerian government at all 
levels. Inputs and financial facilities should be available 
at the appropriate times and in the appropriate quantities 
through agricultural cooperatives. This is because, in the 
past, most of  the Nigerian government’s agricultural 
operations had failed owing to insufficient money and input 
supply. Furthermore, the participation of  ghost farmers in 
such programmes would be confirmed and prevented by 
utilising legal, pre-existing farmers’ cooperatives. This will 
raise agricultural output, reduce food inflation, and increase 
agriculture’s contribution to national GDP. 

(iv) Interest rates should not be permitted to fall below 
a specific level in order to allow for the influx of  both 
local and international direct savings and investment in 
the agricultural sector. This would assist to maximise the 
sector’s potential via competition, which will encourage 
agricultural enterprises to list on the Nigerian stock 
market (NGX) and strengthen the relationship between 
agriculture and manufacturing in Nigeria.  



Pa
ge

 
32

https://journals.e-palli.com/home/index.php/ajase

Am. J. Appl. Stat. Econ. 3(1) 24-32, 2024

REFERENCES
Adekunle, W., Kehinde, A. T., & Taiwo, H. O. (2019). The 

impact of  exchange rate dynamics on agricultural 
output performance in Nigeria. Jos Journal of  Economics, 
8(3), 104 – 128. 

Akinbode, S. O., & Ojo, O. T. (2018). The effect of  
exchange rate volatility on agricultural exports in 
Nigeria: An autoregressive distributed lag (ARDL) 
bounds test approach. Nigerian Journal of  Agricultural 
Economics, 8(1), 11 – 19. http://dx.doi.org/10.22004/
ag.econ.280316

Alegwu, F. O., Aye, C. G., & Asogwa, C. B. (2018). Effect 
of  real exchange rate volatility on agricultural products 
exports in Nigeria. Agris On-line Papers in Economics and 
Informatics, 10(3), 1 – 15. http://dx.doi.org/10.22004/
ag.econ.281641

Awolaja, O. G. & Okedina, I. M (2020) Investigating the 
asymmetric effect of  exchange on agricultural output 
in Nigeria, 1981 - 2017. Central Bank of  Nigeria, 
Economic and Financial Review, 58(4), 41 – 62. https://
www.researchgate.net/publication/366345250

Awoyemi, T. T., & Adeyemi, O. (2018). Exchange rate 
volatility and agricultural sector performance in 
Nigeria. International Journal of  Economics, Commerce and 
Management, 6(2), 73-86. 

Central Bank of  Nigeria (2021).  Statistical bulletin, 
Abuja Nigeria. https://www.cbn.gov.ng/documents/
Statbulletin.asp

Eniekezimene, A. F., Oluwabusayo, A.G., & Ekiye, E. 
(2020). Macroeconomic policies and agricultural 
performance in Nigeria. Nigerian Defence Academy 
Journal of  Economics and Finance, 4(2), 62 – 72.

Gatawa, N. W., & Madmud, A. A. (2017). Impact of  
exchange rate fluctuations on agricultural exports 
(crops) in Nigeria. International Journal of  Humanities 
and Social Science Invention, 6 (3), 65 – 71.

Ikpesu1, F., & Okpe, A.E. (2019). Capital inflows, 
exchange rate and agricultural output in Nigeria. 
Future Business Journal, 5(3), 1 – 8. http://dx.doi.
org/10.1186/s43093-019-0001-9

International Monetary Fund (2021). Foreign Direct 
Investment (FDI). Retrieved from https://www.imf.
org/en/Topics/fdi/overview

Kafle, K., & Kennedy, I. (2015). Exchange rate volatility 
and agricultural trade flows: The case of  the united 
states and OECD countries. Journal of  International 
Agricultural Trade and Development, 9(2), 211 – 231. 
https://doi.org/10.22004/ag.econ.119736

Kipruto, T.A., & Nzai, C. (2018). The effect of  government 
expenditure on agricultural output performance in 
Kenya. International Journal of  Economics, Commerce and 
Management, 6(10), 190 – 202. https://ijecm.co.uk/
wp-content/uploads/2018/10/61012.pdf

Mashinini, S.M., Dlamini, G.S., & Dlamini, G.V. (2019). 
Effect of  monetary policy in Eswatini. International 
Journal of  Economics and Financial Research, 5(5), 94 – 99. 
doi.org/10.32861/ijefr.55.94.99 

Ochalibe, A. I., Okoye, C. U., & Enete, A. A. (2019) 
Analysis of  exchange and interest rate policy 
instruments’ dynamics on agricultural growth in 
Nigeria. International Journal of  Environment, Agriculture 
and Biotechnology, 4(4), 1131 – 1140. https://dx.doi.
org/10.22161/ijeab.4437

Onwuka, O., & Oyewumi, O. (2019). Exchange Rate 
Fluctuations and Agricultural Export Performance 
in Nigeria. Journal of  Agricultural and Environmental 
Sciences, 18(2), 1-14.

Onyeagocha, S. U., Chukwu, K. O., & Nweke, O. C. 
(2021). Exchange rate fluctuations, credit accessibility 
and agricultural sector output in Nigeria. Journal of  
Business and Economic Analysis, 4(1), 1-15.

Reuben, R., & Alala, O. (2014). Impact of  exchange 
rate volatility on Kenya’s tea exports. International 
Journal of  Economics, Commerce and Management, 
2(12), 1 - 18.  https://www.researchgate.net/
publication/304621931

SiriKul, T., Chanchai, B., & Somchai, R. (2015). The effect 
of  exchange rates on agricultural goods for export: A 
case of  Thailand. Information Management and Business 
Review, 7(1) 1 – 11.  https://doi.org/10.22610/imbr.
v7i1.1133

Wagan Z.A., Chen Z., Seelro H., & Shah M.S. (2018). 
Assessing the effect of  monetary policy on agricultural 
growth and food prices. Agric. Econ. – Czech, 64, 499–
507. https://doi.org/10.17221/295/2017-Agricecon


