EFFECT OF MACROECONOMIC VARIABLES ON AGRICULTURAL OUTPUT IN NIGERIA
DOI:10.30825/5.EJPAU.181.2019.22.4

Enoma Ojo, Musa Samuel Olayinka
Department of Economics, University of Lagos, Nigeria

ABSTRACT

Nigeria has been a nation seeking to step-up economic growth and development particularly since the end of the civil war but this appears to have proved too difficult. Agriculture has been earmarked as a key sector to cause faster growth through sufficient food production. However, the output supply from the sector to the domestic economy has been inadequate, leading to massive food importation. This study examined the effect of macroeconomic variables on agricultural output in Nigeria spanning over 37 years, from 1981 to 2017. It is assumed that appropriate macroeconomic variables would serve as transmission channels to address the short comings in agricultural sector. To achieve the study specific objectives, the OLS method was employed as estimation technique. Error Correction Model was set for the short run effects. This was followed by the long run effect of macroeconomic variables on agricultural output in Nigeria. The Granger Causality Test was also carried out. The short run error correction term revealed a speedy adjustment process to long run equilibrium. There was evidence of long run relationship between macroeconomic variables set for this study and agricultural output. The major findings revealed that in the long run, government support in form of microcredit was an essential factor in agricultural development in Nigeria. Exchange rate was also found to be significant in determining agricultural output growth. It is suggested that government should increase its financial support to agriculture and that the stability of the exchange rate should be taken seriously by the apex bank.

Key words: Macroeconomic Policy, Agricultural Output, Cointegration, Error Correction Model, Exchange Rate, Nigeria.

INTRODUCTION

Agriculture and macroeconomic policy have symbiotic relationship. While agriculture is the mainstay of any developing country, macroeconomic policy can be a catalytic factor driving agriculture. Agriculture is an integral part of the macro economy and it requires sound macroeconomic policy for its growth. Macroeconomics is the "big picture" of an economy's overall performance Odior [19]. This means that while agricultural sector might be a channel to achieve macroeconomic policy objectives, macroeconomic policy, on the other hand, can set the path to develop agricultural sector.

Looking conscientiously at the Nigerian macro economy, the growth rate of the various macroeconomic variables such as interest rate, inflation rate, exchange rate and unemployment rate, together with the resultant Gross Domestic Product (GDP) has been asymmetric over the years to the detriment of the aggregate economy particularly agriculture Onakoya, Aroyewun-Khostly and Jonhson [23]. Most economists believe that macroeconomic policy changes often have significant impacts on agricultural economy worldwide. Although policymakers try to design policies to improve the national economy, these policies often have unintentional and damaging consequences on the agricultural sector Odior [19].  

In Nigeria, agricultural sector has been invaluable in supporting economic growth and development since 1960. In fact before the discovery and commercial exploration of petroleum in Nigeria, the country depended on revenue generated from agriculture export. Earnings from this formed critical integral fraction of revenue employed for the development of other sectors of the economy. As a result, the sector has continued to be a target of government policies over time, Eyo [7]. Generally, for countries in the early or medium stage of development, agriculture remains a critical sector to be developed because of its crucial role in food security and employment of labour, Fashola [8]; Musa [15]. In the advanced countries and emerging economies, agriculture is an important source of foreign exchange earnings particularly in the early stage of economic growth.

Nevertheless, in Nigeria, the sector has been characterized by years of inertia and unpredictability in production and marketed volume, even though it is one of the leading sector of the economy contributing about 75% to its non-oil exports, Obasaju and Baiyegunhi [18]. In fact, from independence in 1960 till the very beginning of oil boom era of early 1970s, agriculture contributed over 70% to the GDP and about 90% of foreign exchange earnings, Nnanna, Alade and Odoko [16]. However, the sector started losing its importance in the economy in the wake of oil prosperity which precipitated Dutch disease and de-industrialization. It is sad to note that this seemingly abysmal state of agriculture persists till today and if it continues, there might be undesirable negative consequences. Poverty, hunger and continuing dependence on foreign food. The continue food importation would aggravate Nigeria’s adverse balance of payments as food importers would mount increasing pressure on the available foreign exchange.

Although, there are quite a lot of work done to attend to agricultural problems in Nigeria, many of these are yet to adequately attend to agricultural problems. Some previous works, to mention a few, include Eyo [7], Odior [19], Abubakar and Ibrahim [1]. None of these appears has painstakingly addressed the sector. Also government policies to improve the sector include agricultural programmes, direct order, extension services and subsidies. However, these have failed to yield optimal outcomes. It might be that the policies, for instance, selective interest rate, agricultural credit schemes, extension services and subsidies are too direct and cannot reach many farmers. Therefore, this study differs by examining a broad view of the economy by employing critical macroeconomic policy instruments that might be significant in offering feasible solution to agriculture in Nigeria. Therefore, the objective is to examine both the short run and long run effects of macroeconomic policy on agricultural output performance in Nigeria. We will also examine the causal relationship between the policy variables and agricultural output.

LITERATURE REVIEW

Because of its importance, agriculture is of interest to development economists. Schultz [24] states that until countries can produce a sufficient amount of food, labor is trapped in agriculture and they cannot begin the process of modern growth. This idea has been carried forward by others, such as Johnston and Mellor [12], Johnston and Kilby [11] and Mellor [14]. Gollin [9] summarizes a general idea of this line of thinking, he maintains that increased agricultural productivity is a key to structural transformation and subsequent economic development in response to productivity shocks. McMillan, Rodrik, and Verduzco-Gallo [13] argue that this structural transformation remains an important source of potential economic growth particularly in the Sub-Saharan African countries.

Nwankwu [17] and World Bank [26] observed that in many developed and developing countries agriculture has been the main source of gainful employment, a source of basic food supply with which the nation can feed its teeming population, a regenerative source of foreign exchange earnings, means of providing the nation’s industries with local raw materials, and as a reliable source of government revenue. Olabanji et al. [20] commented that in general, agricultural sector contributes to the development of an economy in four major ways – product contribution, factor contribution, market contribution and foreign exchange contribution. It then follows logically that agricultural development deserves a greater attention in a developing economy like Nigeria. In this study, we review quite a variety of literature employed macroeconomic policy to address the sector. 

The relevant theoretical views considered in the study are the Mundell-Fleming and high-pay-off input models. The basic tenets of high-pay-off input model is the ability to annex current agricultural resources which is assumed to come from outside the agricultural sector itself. It is postulated that factors leading to agricultural development should be exogenous and accompanied by policy, education and also technology which peasants can use and must be affordable. The Mundell-Fleming model explains the linkage between macroeconomic policy instruments and the real sector of which agriculture is one. It can be deduced from the model that policy instruments such as interest rate and exchange rate can serve as effective transmission channels through which government policy reaches the real sector and these instruments, if properly administered, can play a major role in improving performances of the sectors.  

Literature abound explaining the relationship between agriculture and macroeconomic policy. A prominent view of the importance of agriculture for development comes from Schultz [24] characterization of the “food problem”; until countries can produce a sufficient amount of food, labor is trapped in agriculture and they cannot begin the process of modern growth. This idea has been carried forward by others, such as Johnston and Mellor [12], Johnston and Kilby [11], and Mellor [14]. Other authors who share this line of thought are Gollin [9], McMillan, Rodrik, and Verduzco-Gallo [13]. The summary of their studies is that agricultural productivity rise is a key to structural transformation and subsequent economic development in response to productivity shocks and that the economies of the Sub-Saharan African countries need structural transformation to achieve economic growth.

Syed, Muhammad & Rana [25] analyzed the impact of agricultural exports on the macroeconomic performance of Pakistan employing a time series data for the period 1972 to 2008. In their long run analysis, the study found a negative relationship between agricultural export and economic growth, while a non-agricultural export was found to have positive relationship with economic growth. On the basis of the empirical results, the study suggested that Pakistan has to embark on structural changes in agricultural exports by converting its agricultural exports into value added products. Although, as suggested in their study, converting agricultural exports into value added products is applicable to the Nigerian economy but their findings, which shows a negative relationship between agricultural export and economic growth, are not applicable to the Nigerian economy.

There are quite a good number of literature examined the relationship between macroeconomic policy and agricultural sector in Nigeria. Some of these are summarized as follow. Odior [19] examined the effect of macroeconomic policy on Nigerian agricultural performance between 1970 and 2012 by adopting one-step dynamic forecast analysis. Similarly, Imoughele and Ismaila [10] analyzed the impact of exchange rate on nonoil exports in Nigeria between 1986 and 2013 using ordinary Least Square technique. While Odior’s findings revealed that real monetary aggregate, technological change introduced overtime and pass level of agricultural sector performance play a crucial role in affecting the agricultural gross domestic product in Nigeria, Imoughele and Ismaila found a little different outcome. They discovered that exchange rate, money supply, credit to private sector and real GDP have significant impacts on the growth of non-oil exports adding though, that appreciation in exchange rate has negative impact on Nigeria’s non-oil exports.

The critical work of Brownson et al. [4] investigated the effect of macroeconomic variables fluctuation on agricultural productivity in Nigeria over the period of 1970 and 2010 using the techniques of cointegration and error correction model (ECM). Similarly, Akpan et al. [3] adopted the techniques of cointegration and error correction model (ECM) to quantify the role of macroeconomic variables on agricultural diversification in Nigeria over the period of 1960 to 2014. Brownson findings showed that in both long run and short run, real exports, real external reserves, inflation rate, and external debt have significant negative effects on agricultural productivity. On the other hand, industrial capacity utilization and nominal exchange rate promote agricultural productivity in Nigeria. Akpan et al., [3] reported that longrun positive drivers of agricultural diversification include inflation rate, viable manufacturing, credit to agricultural sector, external reserves, per capita income, unemployment and energy consumption. However, crude oil prices, lending capacity of commercial banks, foreign direct investment in agriculture, and non-oil imports constitute negative long-run drivers of the Nigerian economy.

Omojimite [22] and Oluwatoyese et al. [21] examined the macroeconomic factors affecting the Nigeria’s agricultural sector between 1981 and 2013 using fully modified OLS and multivariate cointegration approach respectively. They found that commercial bank loan to agriculture, interest rate and food imports are significant factors affecting agricultural output, whereas exchange rate, inflation rate and unemployment rate turned out to be insignificant factors driving Nigeria’s agricultural output.

In some recent literature like Obasaju and Baiyegunhi [18], Abubakar and Ibrahim [1], Onakoya et al. [23], the effect of macroeconomic variables were examined on agricultural sector over a period of 36 years with time series data ranging from 1980–2016. Obasaju’s study assessed the short-run and long-run linkages of macroeconomic policies with real agricultural output in Nigeria (1980:Q1 – 2014:Q4) by a vector error correction model (VECM) and Variance Decomposition techniques. His study showed that in the long-run, inflation and money supply are the two macroeconomic variables that are statistically significant in explaining variation in real agricultural output, with inflation having greater impact on real agricultural output. Also among his findings are that inflation rate Granger-causes real agricultural output in Nigeria. He recommended the need for policy makers to administer considerate rate of interest while also keeping the price level stable in order to raise effective demand and investment and consequently boost real agricultural output in Nigeria. In the study of Onakoya et al. [23], a critical long run and short run analysis of value addition in agriculture was the main objective. The study covered 1970 to 2016 using vector error correction model (VECM). Their finding show that in the long run, inflation rate, exchange rate and agricultural employment rate were positively related and significant in forecasting the value added in agricultural output. They suggested continuous diversification of the Nigerian economy to facilitate profitable commercial agriculture.    

The empirical review thus showed that most previous authors employed VAR and VECM while others used cointegration technique. This study carry out a long run and short run analysis, this study will add more by examine causal relationships and incorporates a wider range of data set. It is anticipated that the result will give feasible suggestions to address the lackluster performance in the agricultural sector in Nigeria.

METHODOLOGY

In this paper we develop a model for Nigerian agricultural sector and used it to estimate the effect of macroeconomic policy instruments on agricultural sector performance. The policy instruments include those omitted in the previous studies. These include monetary policy rate and economic openness. Other instruments are exchange rate movement, government expenditure in terms of number of loan guaranteed under agricultural credit guarantee scheme fund (ACGSF) and amount of fund allocated to the (ACGSF). It is theorized that output in the agricultural sector is a function of the policy instruments identified above. Time series data ranging from 1981 to 2017 are employed for measurement and coefficient estimation.

The models developed in this study follow previous authors’ literature and theoretical views. It is basically an augmented version of Keynesian model which relates policy instruments with the real sector. The model specification relates with the Mundell-Fleming and high-pay-off input model. It is postulated that factors leading to agricultural development should be exogenous and accompanied by policy instruments that enhance low cost of inputs and boost output volume. With respect to this, given the augmented Cobb-Douglass production function whereby policy instruments are added as inputs in production process:

(1)

Where G represents macroeconomic variable, = proportion of policy inputs, that is, monetary, fiscal and trade policies and , proportion of private capital inputs.  is the privately provided embodied capital broader than capital-labour ratio in equation in the traditional production function, then, equation (1) can be re-written in short run and long run form. Taking the log-linear form of equation (1), yields equation (2) below:

(2)

Assuming Y implies agricultural sectors and G & k in equation (2) represents macroeconomic variables, equation (2) can be augmented to a linear equation for the purpose of estimation of variables specified in the study.

The long run form can be written as follow:

(3)

Where:
AGRIC = Agricultural Output
MPR = Monetary policy rate
OPEN = Economic openness
EXR = Exchange rate movement
NACGSF = Number of loan guaranteed under the agriculture credit guaranteed scheme and fund
LACGSF = Amount of loan given under the agriculture credit guaranteed scheme and fund 
μ = Error term.

All variables are expressed in logarithm form.

The short run error correction model is specified in equation (4) below:

(4)

Where:
ECT = Error term
Δ = Difference operator
x = set of deterministic variables like the constant term and trend
δ0 = vector of coefficients of deterministic variables
Δ = first-difference operator   
r =  optimal lag length
μt = the residual term.

The economic relationship between dependent variable (AGRIC) and independent variables are can be explained in terms of their economic importance and a-priori expectations. Monetary policy rate (MPR) is expected to have inverse relationship with the real sector. In other words, as monetary policy rate, which is the cost of borrowing from the apex bank rises (or falls), the cost of borrowing also falls and existing or intending farmers borrow more or are encouraged to borrow, ceteris paribus, and invest in other to boost production in the next period. Economic openness (OPEN) is defined as the ratio of export plus import divided by GDP. It is assumed the more an economy open, the more is the easiness of doing business. Such economy might attract foreign investor and inflow of foreign capital. The exchange rate (EXR) generally measures the price of foreign commodity, the value of domestic currency and purchasing power of economic agents. A high exchange rate means more of domestic currency would go for foreign currency. A low exchange rate is desirable because it brings about low cost of inputs. Therefore, a fall in exchange rate leads to increase in agricultural output via decrease in cost of inputs. Number of loan guaranteed under the agricultural credit guaranteed scheme and fund (NACGSF) and amount of loan given (LACGSF) are government special assistance to boost agricultural sector in Nigeria. When the loan is given to farmers, it increases their output capacity add more volume to the GDP. Increase in agricultural output keep price stable and increases the nation self-reliance. It also facilitates foreign exchange conservation.

The simple OLS technique is employed to examine the relationship between macroeconomic policy and agricultural sector in Nigeria. Estimation technique is based on error correction model (ECM). It is started by testing for stationarity of the variables and check for co-integration. The existence of co-integration among macro-economic variables and agricultural output implies there is an adjustment process which prevents errors in the long-run relationship from becoming larger and larger.  By this, it is possible not only to capture both shortrun and longrun dynamics but also it, allows for models to be tested for data admissibility and theory consistency Adebiyi [2].  The Error Correction Model popularized by Engle and Granger Correct [5] will be used for the adjustment. According to the authors the co-integrated series, are the necessary condition for an error correction model (ECM) to hold. The ECM incorporates both the economic theory relating to the long run relationship between variables and short-run disequilibrium behaviour. From the model, we will also examine the causal effects of the variables. For this, the Granger causality tests will be used. Data are sourced from the Central Bank of Nigeria (CBN) Statistical Bulletin, 2017.

RESULTS

Table 1. Descriptive Statistics of the Variables Employed for the Regression Analysis

	AGRIC	MPR	OPEN	EXR	NACGSF	LACGSF
Mean	3.023375	1.093247	0.311622	1.556866	4.289662	5.761491
Median	3.154416	1.113943	0.350000	1.966275	4.374400	5.391081
Maximum	4.379352	1.414973	0.590000	2.491362	4.859270	7.113843
Minimum	1.231780	0.778151	0.070000	0.041400	3.031812	4.391903
Std. Dev.	1.057699	0.143250	0.126744	0.811791	0.506684	0.925300
Skewness	-0.322965	-0.400819	-0.220909	-0.843388	-1.272032	0.161533
Kurtosis	1.651531	3.203982	2.453797	2.215277	3.731614	1.466527
Jarque-Bera	3.446541	1.054857	0.760877	5.335709	10.80326	3.786195
Probability	0.178481	0.590120	0.683562	0.069401	0.004509	0.150605
Sum	111.8649	40.45015	11.53000	57.60405	158.7175	213.1752
Sum Sq.  Dev.	40.27421	0.738743	0.578303	23.72415	9.242227	30.82246
Observations	37	37	37	37	37	37
Source: Author’s Computation based on the data obtained from the CBN 2017

Table 1 presents the descriptive statistics of both dependent and independent variables. The mean value of agricultural output, monetary policy rate, economic openness, exchange rate, number of loan guarantee and the amount of loan given under ACGSF are approximately 3.02, 1.09, 0.31, 1.55, 4.29 and 5.76 respectively. The standard deviation for the variables are relatively low except agricultural output and amount of loan given under the scheme which are relatively high at 1.06 and 0.93 respectively. We might infer that in Nigeria, agricultural output yields vary directly with human effort and weather condition which are very unpredictable. As a result, there is a wide range of gap in annual data recorded. The result also shows some level of discrepancies in the amount of loan given each year by the CBN under the LACGSF. This might be as a result of the apex’s bank effort at maintaining macroeconomic stability. This verse difference is also explained in the data range for the two variables. Volatility of exchange rate and economic openness is also displayed in the result. There should be a concerted effort to take cognizance of the variables while formulating policy to address agricultural performance in Nigeria.      

Table 2. Pair-wise Correlation Matrix

	AGRIC	MPR	OPEN	EXR	NACGSF	LACGSF
AGRIC	1
MPR	0.0438	1
OPEN	0.6048	0.3339	1
EXR	0.9588	0.2300	0.7558	1
NACGSF	0.8250	0.2795	0.5708	0.8402	1
LACGSF	0.9533	-0.1212	0.4300	0.8501	0.8160	1
Source: Author’s Computation based on the data obtained from the CBN 2017

The correlation matrix of the variables is presented in Table 2. The Pearson product moment correlation coefficients of the variables are significant at 5% level. There is a weak positive correlation between monetary policy rate and agriculture. This means that MPR may not necessarily be a key variable to address agricultural output performance in Nigeria. All other variables have high positive correlation with agriculture, while exchange rate carries the highest positive correlation of 96%. In other words, policy makers should consider the importance of exchange rate effect on the real sector and in particular, agriculture. In addition, the result also shows that there is a low negative correlation between LACGSF and MPR.

The descriptive analysis cannot be totally rely upon. There is a need to carry out empirical analysis which is expected to generate a scientific and more reliable outcome.  

Table 3. The OLS Initial Regression Results

Dependent Variable: AGRIC Sample: 1981 2017 Included observations: 37
Variable	Coefficient	Std. Error	t-Statistic	Prob.
C	-0.509376	0.362372	-1.405671	0.1698
MPR	0.029332	0.180389	0.162603	0.8719
OPEN	-0.375403	0.258229	-1.453762	0.1561
EXR	0.858167	0.079709	10.76627	0.0000
NACGSF	-0.256586	0.075405	-3.402785	0.0019
LACGSF	0.587050	0.065228	8.999922	0.0000
R-squared	0.992737	Mean dependent var		3.023375
Adjusted R-squared	0.991565	S.D. dependent var		1.057699
S.E. of regression	0.097140	Akaike info criterion		-1.677941
Sum squared resid	0.292519	Schwarz criterion		-1.416711
Log likelihood	37.04191	Hannan-Quinn criter.		-1.585845
F-statistic	847.4193	Durbin-Watson stat		1.276817
Prob(F-statistic)	0.000000

Table 3 above shows the result of the Ordinary Least Squares (OLS) initial regression. According to Yule [27], Engle and Granger [6], running a regression on time series data without testing for the unit root might generate nonsense and spurious results, thereby making our estimation bias and unreliable. Therefore the study proceed to the next stage which includes testing for the unit root. 

Table 4. Augmented Dickey-Fuller (ADF) and Philips-Perron (PP) Unit Root Tests

VARIABLES	ADF	ADF	Order of Integration	PP	PP	Order of Integration
	Level	1st-Difference		Level	1st-Difference
Agric	-1.8914	-3.8322	l(1)	-1.8914	-3.8830	l(1)
MPR	-3.0021	-7.2373	l(1)	-2.9903	-7.3976	l(1)
OPEN	-2.4437	-7.8240	l(1)	-2.2635	12.6740	l(1)
EXR	-1.7421	-5.7710	l(1)	-1.8224	-5.7819	l(1)
NACGSF	-3.4475	3.2541	l(1)	-2.4585	-3.1938	l(1)
LACGSF	-1.1373	3.9164	l(1)	-0.7588	-3.9164	l(1)
Source: Author’s Computation) SIC = Schwarz Information Criterion.    AIC = Akaike Information Criterion

Table 4 shows the level and first difference stationarity test results. The table reveals that all the variables have unit root but using both augmented Dickey-Fuller (ADF) and Philips-Perron (PP) unit root tests, they are stationary at first difference. The study proceeds to the co-integration test for long run relationship.

Table 5. Johansen Co-Integration Test

Date: 04/22/19   Time: 13:32
Sample (adjusted): 1983 2017
Included observations: 35 after adjustments
Trend assumption: Linear deterministic trend
Series: AGRIC MPR OPEN EXR NACGSF LACGSF
Lags interval (in first differences): 1 to 1
Unrestricted Cointegration Rank Test (Trace)
Hypothesized		Trace	0.05
No. of CE(s)	Eigenvalue	Statistic	Critical Value	Prob.**
None *	0.743825	120.3615	95.75366	0.0004
At most 1 *	0.534589	72.69523	69.81889	0.0289
At most 2	0.434340	45.92605	47.85613	0.0751
At most 3	0.308181	25.98436	29.79707	0.1292
At most 4	0.183824	13.08928	15.49471	0.1116
At most 5 *	0.157055	5.979892	3.841466	0.0145
Trace test indicates 2 cointegrating eqn(s) at the 0.05 level  * denotes rejection of the hypothesis at the 0.05 level  **MacKinnon-Haug-Michelis (1999) p-values

From Table 5, there are at least two co-integrating equations. This means that the independent variables jointly have long run relationship on agricultural output in Nigeria. That is, they can be employed simultaneously to address agriculture.  

Table 5.1. Normalized Co-integrating Coefficients (standard error in parentheses)

AGRIC	MPR	OPEN	EXR	NACGSF	LACGSF
1.000000	-0.386074	-1.648682	-0.513508	0.621652	-0.938345
	(0.20273)	(0.32534)	(0.08569)	(0.07943)	(0.07572)

We interpret the normalized co-integration result in reverse form, therefore, from Table 5.1, all the variables are positively related with agriculture except number of loans given which is negative. The variables are essential to address the long run effects of macroeconomic policy on agriculture in Nigeria.

Table 6. Empirical Result of the Short Run Effect of Macroeconomic Variables In Agricultural Output in Nigeria. Equation (3.9), Specific Objective I Dependent Variable = Agriculture

Variable	Coefficient	Std. Error	t-Statistic	Probability
D(MPR(-1))	-0.013532	0.122950	-0.110064	0.9131
D(OPEN(-1))	-0.100482	0.177390	-0.566445	0.5756
D(EXR(-1))	0.170760	0.138158	1.235977	0.2267
D(NACGSF(-1))	0.010757	0.098520	0.109184	0.9138
D(LACGSF(-1))	0.148590	0.117325	1.266486	0.2158
ECT(-1)	-0.643507***	0.167721	-3.836770	0.0001

R-Squared	0.7643	F-Statistics	1.9180 (0.0967)
Adjusted R-Squared	0.6946	D.W Statistics	1.4387

*** = Significant at 1% level; ( ) Probability value

The short run effect of macroeconomic variables on agricultural output in Nigeria is presented in Table 6. In the short run, economic openness and monetary policy rate are negatively related with agriculture. However, other variables are positive though the relationship are weak or insignificant. A unit change in macroeconomic variables will result in a minute change in agricultural output. That is, in the short run, macroeconomic variables might have trivial effects on agriculture. This may be as a result of the lag in agricultural response to policy adjustment or poorly designed policy.  

The more government opens the economy, the more there is a decline in agricultural performance. This suggests that in the short run, economic openness might appear detrimental to the Nigerian agricultural sector. Monetary policy rate conforms with a-priory expectation. A fall in MPR, holding other variables constant, will lead to a rise in agricultural output. Following Keynesian postulation, a fall in interest rate can induce investment in capital goods, especially, those used by farmers. Exchange rate does not conform with a-priori expectation. A fall in exchange rate should induce a rise in agricultural performance. The ambiguity of exchange rate result in the short run might be the result of the seemingly large operation of informal sector or black market activities prevalent in the Nigerian money market. The number of loan given under credit scheme and the amount of loan given are positively related with agricultural output. The higher the credit given to the sector, the greater the output. 

The R-Squared (0.7643) and Adjusted R-Squared (0.6946) clearly indicate a strong relationship between macroeconomic variables and agricultural output. At least, 69% of the changes in agricultural output is determined by the macroeconomic variables. The Durbin Watson statistics (1.4387) does not fall within the acceptable region in statistics, indicating present of positive autocorrelation. This means that if a macroeconomic variable is augmented to economically advantageous position in the present period, there would likely be an increase in the value of agricultural output in the next period    

Following the objective 1, a short run positive relationship exists between macroeconomic variables and agricultural output in Nigeria, but the fact that none of the variable is significant in the relationship shows that we cannot rely on macroeconomic variables alone as short run policy variables to address agricultural sector in Nigeria.    

The negative term of the error correction model (-0.6435) indicates the adjustment to equilibrium process is fast. The speed of adjustment to long run equilibrium is about 64%. In other words, macroeconomic variables and the performance of agriculture tend to affect one-another in the long run. Moreover, any short run disequilibrium can be corrected in the long run.    

Table 7. Empirical Result of the Long Run Effect of Macroeconomic Variables In Agricultural Output in Nigeria. Equation (3.8), Specific Objective II Dependent Variable = Agriculture

Variable	Coefficient	Std. Error	t-Statistic	Probability
MPR(-1)	0.109239	0.192012	0.568919	0.5736
OPEN(-1)	-0.374035	0.269836	-1.386159	0.1759
EXR(-1)	-0.860639***	0.083287	-10.33344	0.0000
NACGSF(-1)	-0.212433***	0.080473	-2.639794	0.0030
LACGSF(-1)	0.536916***	0.069117	7.768222	0.0000

R-Squared	0.8916	D.W Statistics	1.1591
Adjusted R-Squared	0.8202	LM	0.2490 (0.0502)
F-Statistics	711.8118 (0.0000)	χ2	0.0353 (0.0440)

***, **, * = Significant at 1, 5 and 10% level; ( ) Probability value

From Table 7, the long run effect of macroeconomic variables on agricultural output shows that exchange rate, the number of loan given under ACGSF and the amount are significant at 1% level. However, the expected sign of NACGSF fail to conform with a-priori expectation. The amount of loan given (LACGSF) is positively related with AGRIC as expected. In the long run, government subsidy in the form of agricultural support loan is an essential factor in agricultural development in Nigeria. One period lag of the MPR and OPEN are not significant and the coefficients, 0.10 and -0.37 respectively do not carry expected signs. The 0.10 for MPR means that a unit increase in monetary policy rate will lead to 10% increase in agricultural output. On the other hand, a negative change of 37% occurs in agricultural output when economic openness changes by a unit.

The R-Squared and the Adjusted R-Squared are both high at approximately 89 and 82% respectively. That is over 82% of the changes occur in agricultural output is explained by the macroeconomic variables used. Durbin Watson statistics (1.1591) does not fall within the acceptable region in statistics, indicating present of positive autocorrelation in the residual of the variables. This means that the behavioral pattern of the residuals in one period might follow the same pattern in the next period.

Sensitivity Analysis
Quite a good number of macroeconomic variables are employed in the process of regression analysis. The most robust variables are chosen for our analysis. The robustness of the variables is apparent from the diagnostic tests. To ascertain the goodness of fit of the long run model, stability diagnostics tests are conducted. The stability test of the regression coefficients is conducted by employing the cumulative sum (CUSUM) and the cumulative sum of squares (CUSUMSq) of the recursive residual test for structural stability. The tests are applied to the residuals of all the short run and long run models. The null hypothesis that the coefficients of the error correction model for the long-run model are not stable is accepted for CUSUM test but rejected for CUSUMSq test. Statistics exceed the bounds of the 5% level of significance.

The diagnostic test examines the serial correlation, functional form, normality and heteroscedasticity associated with the model. From both the short run and long run diagnostic tests, it is clear that there is no serial correlation among variables because functional form of model is well specified. The Breusch–Godfrey serial correlation Lagrangian multiplier (LM) test is employed to achieve this. Autoregressive conditional heteroscedasticity (ARCH) is present in the models following the ARCH tests which suggest that the errors are heteroskedastic and are not independent of the regressors. Jarque-Bera (JB) test is engaged in order to test the normality of error term. All the models pass the normality tests, suggesting that the errors are normally distributed.

Table 8. Presentation of the Result of Granger Causality Test, Specific Objective III

Pairwise Granger Causality Tests
Date: 04/22/19   Time: 13:35
Sample: 1981 2017
Lags: 2
Null Hypothesis:	Obs	F-Statistic	Prob.
LACGSF does not Granger Cause AGRIC	35	1.80380	0.1821
AGRIC does not Granger Cause LACGSF		2.49318	0.0996
OPEN does not Granger Cause MPR	35	0.25732	0.7748
MPR does not Granger Cause OPEN		7.63606	0.0021
EXR does not Granger Cause MPR	35	1.40444	0.2612
MPR does not Granger Cause EXR		6.19367	0.0056
NACGSF does not Granger Cause MPR	35	0.05661	0.9451
MPR does not Granger Cause NACGSF		0.57609	0.5682
LACGSF does not Granger Cause MPR	35	1.29166	0.2897
MPR does not Granger Cause LACGSF		0.27700	0.7600
EXR does not Granger Cause OPEN	35	1.54240	0.2304
OPEN does not Granger Cause EXR		1.33577	0.2781
NACGSF does not Granger Cause OPEN	35	1.94440	0.1607
OPEN does not Granger Cause NACGSF		0.34133	0.7136
LACGSF does not Granger Cause OPEN	35	0.08993	0.9142
OPEN does not Granger Cause LACGSF		0.23819	0.7895
NACGSF does not Granger Cause EXR	35	1.15303	0.3293
EXR does not Granger Cause NACGSF		0.18247	0.8341
LACGSF does not Granger Cause EXR	35	0.40982	0.6674
EXR does not Granger Cause LACGSF		1.17862	0.3215
LACGSF does not Granger Cause NACGSF	35	1.10257	0.3451
NACGSF does not Granger Cause LACGSF		0.76756	0.4730

From Table 8, given the F-statistics (2.49318) and probability value (0.0996), significant at 10%, the null hypothesis is rejected. The alternative hypothesis is accepted and we conclude that there is a unidirectional causality running from agriculture (AGRIC) to amount of loan given under the agricultural credit guarantee scheme and fund (LACGSF). In other words, agriculture can cause or induce more loan from the government. As agriculture expands, there is the tendency government expenditure increases thereby increases the macroeconomic responsibility of the government. Also, the null hypothesis is rejected for the causal relationship between monetary policy rate (MPR) and economic openness (OPEN); and the MPR and exchange rate (EXR). That is a unidirectional granger causality running from MPR and the two variables. Also, the relationship is significant at 1%. Note that the result does not show bidirectional relationship between any two variables. That is, where causal relationship exists, it is unidirectional. Moreover, in most instances, pairing the variables does not show significant causality. This suggests poor implementation of policy and lopsided macroeconomic framework in Nigeria.

We find evidence of existence of short run and long run relationships between agriculture and macroeconomic variables. This is established in the co-integration test. Although the variables are not significant in the short run, but they are co-integrated to establish a long run relationship. The significance of the error correction also establishes speedy adjustment process, which means that any short run error can be corrected in the long run with minimum lag effect. That is policy formulated based on the model developed in this study would yield desirable objective in the long run.

This study found that in the long run, loan given to support agriculture is growth inducing in the sector but this may also have implication on government revenue. Government has alternative uses for revenue generated. Infrastructural provision and debt services also compete for government income. Therefore, government can only allocate a fair share to agricultural sector which in most cases may not be enough. With respect to this, private sector participation might be a sort of additional value.

Monetary policy rate and economic openness are not necessarily essential as macroeconomic variables to adequately address agricultural sector. Nigerian financial system appears to be underdeveloped and this probably resulting in failure to achieve positive short run objective on interest rate adjustment. Economic openness which sometimes involves liberalization and free mobility of capital is not necessarily effective to address long run agricultural performance in Nigeria. This might be as a result of the fact that agricultural activities are primitive and often there is no significant trans-border transaction with direct linkage to agricultural output performance (vis-à-vis macroeconomic policy instruments). With reference to exchange rate, the long run result proves exchange rate to be economically relevant as a policy variable. A stable rate is crucial for successful implementation of policy because the variable affects prices and input in the domestic economy.

One may conclude that the results of the findings in this study are relevant to agricultural sector and can suggest way forward to finding solution to agriculture slow growth in Nigeria.

DISCUSSION

The findings obtained in this study highlight the importance of macroeconomic variables in agricultural development in Nigeria. To achieve the study specific objectives, first, the OLS method was employed to estimate the error correction model set for the short run effects. Secondly, the study estimated the long run effect of macroeconomic variables on agricultural output in Nigeria. Lastly, Granger Causality Test was also carried out. The short run error correction term reveals a speedy adjustment process to long run equilibrium. That is, any short run in policy lag can be corrected within a short term frame and that there is possibility of policy achieving its objectives in the long run. There is evidence of long run relationship between macroeconomic variables set for this study and agricultural output. The Granger causality result shows unidirectional causality between few variables. No bi-directional causal relationship obtained. Nevertheless, the major findings revealed that in the long run, government support in form of microcredit is an essential factor in agricultural development in Nigeria. That is, government financial support to agriculture is important in Nigeria to boost productivity and output. The evidence is shown in the result that the number of loan given under credit scheme and the amount of loan given are positively related with agricultural output. The higher the credit given to the sector, the greater the output.

Exchange rate is also found to be significant in determining agricultural output growth. There is, however, quite some insignificant evidence of volatility in exchange rate effects in the short run which might have arisen as a result of quite obvious operation of the informal sector or black market in Nigeria. Therefore, there is hope for the Nigerian agriculture if a concerted effort is made to focus on the macroeconomic variables discover as significant in this study. This then leads to the conclusion of the study

CONCLUSIONS

This study basically examined the effect of macroeconomic variables on agricultural output in Nigeria spanning over 37 years, from 1981 to 2017. Nigeria has been a nation seeking economic growth and development since independence. This desire has proved too difficult to achieve. Government has made several efforts to address the slow growth rate in the economy through multi-sectoral development. Agriculture has been one of the major target sectors. Although the sector has received lots of support from the government, yet the much output required by the society are not produced. Food production to support the economy has been inadequate. Quite a handful of policies have been adopted to address the sector but with no significant result. This study therefore examines macroeconomic variables effect on the sector. The result obtained from the research proved that government still needs greater role to play concerning agricultural development in Nigeria. There is hope for Nigeria in food production but appropriate macro economy must be established to broaden the channels through which financial resources flow into the agricultural sector. Finally, from the results obtained in the course of this study, the following policy recommendations are suggested:

Firstly, loan support to agriculture is very important in Nigeria. With respect to this, budgetary allocation to agricultural sector should be increased significantly so that adequate funds can be available for driving the activities of the sector. Budgetary implementation in the agricultural sector should also be pursued so as to foster a higher level of budget implementation in other areas, such as for capital projects. This will ultimately ensure that the agriculture food production attainment which is geared towards achieving food security, poverty reduction, employment generation and wealth creation, is realized in Nigeria.

Moreover, for the Nigerian economy, the role of government is vital because government is the major driver of the economy through the size of its income and expenditure. It can catalyze the economy to the desired outcomes through its fiscal and monetary policies actions. Therefore, increasing government expenditure on agricultural programmes, extension services and subsidies are additional key ingredients in agricultural output growth. Following these, monetary policy action such as concessionary interest rate for farmers and also supply-side policy like strict implementation of the Land Use Act can be added advantages.

Stability of exchange rate is also very crucial in agricultural output growth process. This enhances commercial farming and facilitates foreign investment. It also induces domestic farmers to practice commercial farming if they are able to import capital inputs at a relatively stable and reasonable exchange rate. Nevertheless, stable macroeconomic framework is critically needed to attain stable exchange rate. The government through the apex bank is critically needed to swing into action in this regard.

REFERENCES

1. Abubakar I.F., Ibrahim U.B., 2019. A macroeconomic analysis of agricultural sector in Nigeria. Advanced Journal of Social Sciences, 5(1), 18–25.
2. Adebiyi M.A., 2002. Debt Service and Education Expenditure Nexus: The Nigerian Experience. A selected paper for 2002 Annual Conference. The Nigeria Economic Society. 243–265.
3. Akpan S.B., Udoka S.J., Patrick I.V., 2015. Roles of Macroeconomic variables on Agricultural Diversification. American Journal of Economics and Business Administration, 7 (2), 77–93.
4. Brownson S., Vincent I., Emmanuel G., Etim D., 2012. Agricultural Productivity and Macroeconomic Variable Fluctuation in Nigeria. International Journal of Economics and Finance, 4(8), 114–125.
5. Central Bank of Nigeria, Statistical Bulletin, 2017.
6. Engle R.F., Granger C.W., 1991. Co integration and error correction, representation, estimation and testing. Econometrical, 55, 251–276.
7. Eyo E.O., 2008. Macroeconomic environment and agricultural sector growth in Nigeria. World Journal of Agricultural Sciences, 4 (6), 781–786. ISSN 1817-3047. IDOSI Publications.
8. Fashola M.A., 2005. Agricultural Development in Nigeria [in:] Issues in Money, Finance and Economic Management in Nigeria, Essays in Honour of Prof. Obasanmi Olakanpo. Edited by Fakiyesi O.O. & Akano O. Department of Economics, University of Lagos Unilag Press. 678–702.
9. Gollin D., 2010. Agricultural productivity and economic growth [in:] P. Pingali, R. Evenson (Eds.). Handbook of agricultural economics, vol. 4, 3825–3866. Amsterdam: Elsevier.
10. Imoughele L.E., Ismaila M., 2015. Impact of Exchange Rate on Non-oil exports. International Journal of Academic Research in Accounting, Finance and Management Sciences, 5(1), 190–198.
11. Johnston B.F., Kilby P., 1975. Agriculture and structural transformation: economic strategies in late-developing countries. New York, NY: Oxford University Press.
12. Johnston B.F., Mellor J.W., 1961. The role of agriculture in economic development. American Economic Review, 51(4), 566–593.
13. McMillan M., Rodrik D., Verduzco-Gallo I., 2014. Globalization, structural change, and productivity growth, with an update on Africa. World Development, 63, 11–32.
14. Mellor J.W., 1995. Introduction [in:] J.W. Mellor (Ed.), Agriculture on the road to industrialization. Baltimore: Johns Hopkins University Press.
15. Musa S.O., 2019. Input requirements and technical know-how: issues in agricultural value chain in Nigeria. Electronic Journal of Polish Agricultural Universities (EJPAU), 22 (1).
16. Nnanna O.J., Alade S.O., Odoko F.O., 2003. Contemporary economic policy issues in Nigeria. Central Bank of Nigeria. CBN publication, 131–184.
17. Nwankwu O.O., 1981. Analysis of the contribution of Agricultural sector on the Nigerian Economic Development, World Review of Business Research, 1(1), 191–200.
18. Obasaju B.O., Baiyegunhi L.S., 2019. Agricultural output and macroeconomic policy: examining the delicate linkage in Nigeria. The Journal of Developing Areas, 53(1). Tennessee State University College of Business.
19. Odior E.S., 2014. The macroeconomic policy effect on Nigerian agricultural performance: one-Step dynamic forecasting analysis. International Journal of Economics and Finance. 6(9), 2014. Pub. Canadian Center of Science and Education.
20. Olabanji E., Adebisi F., Ese U., Emmanuel O., 2017. Agricultural Output and Economic Growth in Nigeria Journal of African Research in Business & Technology, (1). IBIMA Publishing.
21. Oluwatoyose O.P., Applanaidu S.D., Abdul Rasak N.A., 2016. Macroeconomic Factors and Agricultural Sector in Nigeria. Procedia Social and Behavioural Sciences, 219, 562–570.
22. Omojimite B.U., 2012. “Institutions, Macroeconomic Policy and the Growth of the Agricultural Sector in Nigeria”. Global Journal of Human Social Science, 12 (1), 1–8.
23. Onakoya A.B., Aroyewun-Khostly B., Johnson B.S., 2018. Value added agricultural output and macroeconomic dynamics in the Nigerian Economy. KIU Journal of Humanities 3(4): 79–91.
24. Schultz T.W., 1953. The economic organization of agriculture. New York, NY, McGraw-Hill.
25. Syed W.A., Muhammad A., Rana M., 2015, “Agricultural Export and Economic Growth: A Case Study of Pakistan”, International Journal of African and Asian Studies, 13, 1–10.
26. World Bank, 2014. World Bank Development Report. Yule G.U., 1926. Why do we sometimes get nonsense correlation between time series? A study in the Sampling and the Nature of time Series. Journal of the Royal Statistical Society, 89, 1–64.
27. Yule G.U., 1926. Why do we sometimes get nonsense correlation between time series? A study in the Sampling and the Nature of time Series. Journal of the Royal Statistical Society, 89, 1–64.

Received: 15.07.2019
Reviewed: 10.12.2019
Accepted: 19.12.2019

---

Enoma Ojo
Department of Economics, University of Lagos, Nigeria

email: enomaojo@gmail.com

Musa Samuel Olayinka
Department of Economics, University of Lagos, Nigeria

email: olayinksam@yahoo.com

---

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.

---

Main  -  Issues  -  How to Submit  -  From the Publisher  -  Search  -  Subscription