Funding the Nigerian Agricultural Sector for Improved Performance: The Role of Migrant Workers’ Foreign Remittances Inflows

Although there is a plethora of empirical evidence on the impact of foreign remittances inflows on economic growth in Nigeria, there are few studies that investigated the impact of foreign remittances on agricultural sector performance in the country. This study therefore examined the impact of funding on the performance of the agricultural sector in Nigeria with emphasis on the role of foreign remittances inflows. Specifically, the study investigated the impact of foreign remittances inflows, government expenditure on agriculture, bank credit to agriculture, the Agricultural Credit Guarantee Scheme Fund, and exchange rate on agricultural sector output performance in Nigeria. Exchange rate was used as a control variable. The Johansen co-integration test, Error Correction Mechanism (ECM), and Granger causality test were used to estimate the annual time-series data for the period of 1981 to 2021. The findings from the study revealed that foreign remittances inflows, government expenditure on agriculture, and loans guaranteed under the Agricultural Credit Guarantee Scheme Fund (ACGSE) have significant positive impact on agricultural sector performance. On the other hand, bank credits to agriculture and exchange rate have insignificant negative impact on agricultural sector performance. Among other things, it is recommended that recipients of foreign remittances in Nigeria should be encouraged to invest more in the agricultural sector.


Introduction
Before independence, and up to the first post-independence decade, agriculture was the main stay of the Nigerian economy.However, the discovery of crude oil in 1956 and the subsequent oil boom of the 1970s practically sniffed life out of the agricultural sector.The oil boom resulted in a shortage of labour in the sector as members of the rural agrarian workforce migrated to the urban centres in search of white-collar jobs.The attention of the government was also diverted from the agricultural sector since its revenue and expenditure needs were met from the proceeds of crude oil exports (Anyanwu et al, 1997, Wilson, 2002).Consequently, the once vibrant agricultural sector was neglected as a result of the oil syndrome.Today, Nigeria is a net importer of food and other raw materials needed in the agroallied industry.There is also a significant reduction in the capacity of the agricultural sector to absorb the country's large workforce.This has resulted in massive unemployment in the country (Olukemi, 2018).
Several factors responsible for the poor performance of the agricultural sector have been identified.These include; low income and poverty of the farmers, low rate of capital formation, lack of modern storage facilities, and inadequate credit facilities, among other things.The farmers are generally poor because of low income.Since the farmers are poor, they could not accumulate large enough capital to enable them procure modern farm inputs that would improve their productivity.Also, the farmers lack the required collateral security with which to acquire large enough credit facilities from financial institutions in the country (Lambo, 1989;Wilson, 2002).
From the foregoing, it is clear that poor funding or lack of capital is a serious problem militating against the development of the agricultural sector in Nigeria.It is also true that due to the preponderance of low income among the farmers, it would be difficult to accumulate enough financial resources required to revamp the agricultural sector from revenue generated internally from the sector.It is important to note that the United Nations Food and Agricultural Organization estimated that developing countries need about USD 83 billion annual investment in the agrarian sector to meet their food requirement (Okpe & Ikpesu, 2019).Meanwhile, it has been reported that Nigeria is among the top recipients of migrant workers' remittances in Africa.Official records also indicate that there are about 1.24 million Nigerian migrants in the diaspora (Nevin & Omosomi, 2019).It is therefore logical to say that the gap in the resources needed to improve the performance of the Nigerian agricultural sector can be bridged through the inflow of migrant workers' remittances.The argument is that, if a reasonable portion of the total annual remittances inflows is channelled to the agricultural sector, it will complement other sources of funding to improve the performance of the sector.
The foregoing discussion provides the background against which this study is designed.The study therefore examined the impact of funding on the performance of the Nigerian agricultural sector with emphasis on the role of migrant workers' remittances inflows.Specifically, the study examined the impact of migrant workers' remittances inflows, government expenditure on agriculture, bank credit to agriculture, and the Agricultural Credit Guarantee Scheme Fund on the performance of the agricultural sector in Nigeria.The paper is structured into five sections.Section one focuses on the introduction.Section two is concerned with literature review.The methodological procedure used in conducting the study is examined in section three.The result of the data analysis and discussion of findings are presented in section four, while conclusions and recommendations are taken care of in section five.

Nigeria's Agricultural Sector
The agricultural sector in Nigeria is divided into four subsectors or components.These are crop production, livestock, fishery and forestry.In Nigeria, the agrarian sector is made up of mixed system of traditional (informal) and modern (formal) farming activities.The National Bureau of Statistics (NBS) estimates show that, on the average, the traditional agricultural system accounted for 90 per cent of the total agricultural output, while the modern farming sector accounts for the balance of 10 per cent.Traditional farming is characterized by production for subsistence, extensive use of land and the practice of shifting cultivation, and a land tenure system based on ownership and access through the family system (CBN, 2010).
Overall, agriculture has remained an important sector in the economy, employing a good percentage of the labour force and contributing significantly to the country's GDP.Specifically, the share of agriculture to total GDP averaged 38.1 per cent, 39.3 per cent and 24.8 per cent for the periods 1981-1989, 1990-1999 and 2000-2010 respectively.The share of agriculture to GDP dropped to 23.9 per cent for the 2011-2019.Analysis of the structure of the agricultural sector by economic activities in 2018 showed that crop production remained the predominant subsector (90.0 per cent), followed by livestock (6.3 per cent), fishery (3.3 per cent) and forestry (1.3 per cent)."In 2020, crop production output was ₦16544.52 billion constituting about 90.17 per cent of the total agricultural output.The livestock subsector contributed ₦1233.11 billion (6.72 per cent) to the total agricultural output.Fishery was ₦380.03 billion representing 2.07 per cent of total agricultural output while forestry was ₦190.51 billion (1.04%).By 2021, crop production stood at ₦16920.52 billion (90.3 per cent); livestock was ₦1240.22 billion (6.62 per cent); fishery contributes ₦384.45 billion (2.05 per cent) while forestry was ₦193.22 billion (1.0 per cent) (CBN, 2021) Theoretical Literature Review 1. Theories of Remittances Theories of remittances are theories that seek to explain the motivating factors that make migrants to send money and other resources to their households in their home countries.These theories include pure altruism, pure self-interest, implicit familyloan agreement, and the intention-to-return theories.
Proponents of pure altruism hypothesis suggest that individual family members are obligated to help each other and that explains migrants' decisions to remit.The theory suggest that a migrant will willingly sacrifice his or her own welfare or interest to take care of the wellbeing of family members due to the love and concern for them (Osili, 2007).The pure self-interest theory, which is also called the tempered altruism or enlightened self-interest theory, was suggested by Lucas and Stark (1985).The theory states that the decision to remit results from the mutually beneficial informal contractual agreement between the migrant and his or her household.According to Lucas and Stark (1985), remittances are claims on implicit co-insurance contracts and because such informal contracts are mutually beneficial, they are self-enforcing and ensure that there is no delinquency.The implicit familyloan agreement theory suggested by Poirine (1997) states that migrants' remittances are the results of family -loan agreements that are entered into by migrants during their youth.Poirine (1997) posits that migrant remittances result from the existence of internal familial markets which finance the investments in human capital of young family members.Therefore, the repayments of such loan contracts appear as remittances when they are made by family members who reside in different countries (Rapoport & Docquire, 2005).The intention-to -return theory is based on empirical studies conducted in Kenya and Botswana by Hoddinott (1994), and Lucas and Stark (1985) respectively.Intention-to-return is often cited as a motive for remitting funds and other resources by migrants to their country of origin.It is ascribed to the motivation on the part of the migrant to return home and inherit what he has invested (remitted) over the years (Collier et al, 2011).

Theories of Agricultural Development
Several theoretical models have been developed to explain societal and environmental forces that lead to agricultural improvement.These models include the conservation, urban-industrial impact, diffusion, and high pay-off input models.However, due to the limitations associated with the conservation, urban-industrial impact and diffusion models, the high pay-off input model is rated the highest among them.In addition, the high pay-off input model encompasses the central themes of the other three models and was formulated to take care of the inefficacy of policy actions based on the other models.In fact, the high pay-off input model is highly ranked due to the desirable impact resulting from its application in Mexico, the Philippines, and other countries in Asia, Africa and Latin America (Moseman, 1970;Udemezue & Osegbue, 2018).
The high pay-off input model developed by Schultz in 1964 is based on a view point developed during the 1960s (Ruttan, 1977).The viewpoint is that farmers in traditional agrarian societies are rational and have the ability to allocate resources efficiently.The farmers, according to this view point, remained poor due to the fact that in most developing countries, there were not enough technical and economic opportunities available for them to explore.Schultz (1964) therefore believe that if high pay-off inputs are made available to the farmers in less developed countries, their primitive agrarian sectors will be transformed into a viable source of economic growth (Moseman, 1970).This study is therefore based on the high pay-off input model of agricultural development.

Empirical Literature Review
Alleluyanatha and Treasure (2021) examined the impact of remittances on crop productivity and household welfare in south-eastern Nigeria using 714 households selected form Anambra and Imo states.The findings showed that there is no significant difference between remittances receiving and non-remittances receiving households.The study also established that remittances have effects on crop productivity.Etim et. al (2020) analysed the impact of remittances on a sample of 120 small holder farmers form six Agricultural Development Project (ADP) zones in Akwa-Ibom state.The result showed that education, experience, farm size and labour have significant positive impact on output of remittance-receiving households.Edeh et al (2020) studied the impact of government spending on agriculture on output performance of the sector in Nigeria.The outcome of the study showed that government capital expenditure on agriculture has significant effect on agricultural output while government recurrent expenditure has insignificant negative impact on agricultural output.Korgbeelo (2019) analysed the impact of public sector spending on agricultural sector performance in Nigeria.The findings showed that government capital and recurrent expenditure have insignificant impact on agricultural sector performance.Ajibade et. al (2018) examined the sources and uses of remittances by farming households in Kwara state, Nigeria.The result showed that age of household head, marital status, gender, number of remitters, etc. are the significant determinants of remittances utilization.Ogburu et.al (2018) examined the impact of government spending on agriculture on unemployment in Nigeria.The study showed that government spending on agriculture has no significant impact on unemployment.Adejumo and Ikhide (2017) investigated the impact of remittances inflows on sectoral performance of the Nigerian economy.The findings showed that remittances have significant positive impact on tradable agricultural, manufacturing and merchandize sector exports.Okoh et. al (2017) examined the impact of migrants' remittances inflows on human capital development and agricultural productivity in Nigeria.The finding showed that remittances have no significant impact on human capital development, while it has significant positive impact on agricultural productivity.Akpan et. al (2017) examined the impact of remittances inflows on indicators of agricultural productivity in Nigeria.The study established that remittances have no significant impact on agricultural GDP/total GDP, agricultural productivity index, and crop production index.Zira and Ezie (2017) investigated the impact of government spending on agricultural production in Nigeria.The findings indicated that government capital expenditure has insignificant positive impact on agricultural production, while government recurrent expenditure has significant positive impact on agricultural production.Nnaji and Nweze (2016) investigated the types, sources and uses of remittances by rural households for agricultural purposes in Enugu state, Nigeria.The study established that 64.4% and 53.3% of the households channelled remittances to crop production and livestock production respectively.Obi and Obayori (2016) in their study established that both capital and recurrent expenditure on agriculture have insignificant positive impact on agricultural production.Okoh (2016) established positive impact of government agricultural expenditure on agricultural productivity in Nigeria.
From the empirical literature reviewed, it is observed that recent studies on the impact of remittance on agricultural sector performance in Nigeria are mostly location -specific.That is, they are case studies of one state or the other.Hence, empirical studies that utilized national data to examine the impact of remittances inflows on the output performance of the agricultural sector in Nigeria are scanty.

Methodology
The study made use of annual time series data for the period 1981 to 2021.The data were obtained from the Central Bank of Nigeria annual statistical bulletin for 2021, the Central Bank of Nigeria annual reports and statements of account (various years) and the World Bank Development Indicators (various years).
The dependent variable for this study is agricultural sector output performance.It is the annual total contribution of the agricultural sector to Nigeria's real GDP.It is measured in billions of naira.The explanatory variables include the foreign remittances, Government expenditure on agriculture, Bank credit to agriculture and Agricultural Credit Guarantee Scheme Fund, while exchange rate is used as a control variable
The model can be expressed in Logarithmic form as follows: Where, L is the natural logarithm of the variables.All other variables are as earlier defined.

Apriori Theoretical Expectations
Based on economic theory, we expect the following signs of the parameter estimates.
>0,  >0,  >0,  >0,  < 0 The implication of the above signs of the parameter estimates is that an increase in remittances inflows, government expenditure on agriculture, bank credit to agriculture and loans guaranteed under the Agricultural Credit Guarantee Scheme Fund will lead to improved agricultural sector performance.On the other hand, an increase in the Naira-Dollar exchange rate will bring about a reduction in the performance of the agricultural sector.

Techniques of Data Estimation
The classical least squares regression technique is based on the implicit assumption that the underlying time-series data are stationary.However, in real life, many macroeconomic time-series are non-stationary (Gujarati & Porter, 2009).
Therefore, due to the problem of non-stationarity associated with time-series data, the analytical procedure was started with stationarity test.To conduct the stationarity test, the Augmented Dickey-Fuller (ADF) unit root test was used.The ADF unit root test is estimated in its general form by the following regressions.
Where  is a time-series, t is a linear time trend, ∆ is the first difference operator,  is a constant, n is the optimum number of lags in the dependent variable, and ∈ is the random error term.
The ADF test relies on rejecting the null hypothesis of unit root (i.e.series is nonstationary) in favour of the alternative hypothesis of no unit root (i.e.series is stationary).
Based on the result of the ADF unit root test, the Johansen cointegration test was used to test whether there exist long-run equilibrium relationships or not among the variables in the model.The essence of the cointergration test is to avoid the problem of spurious regression results.Johansen (1988) and Johansen and Juselius (1990) suggested the test which is based on the Vector Autoregressive (VAR) model.The test starts with a p-lag VAR model specified as follows: =   +   + ….. +   +  + ∈ ……………………....... 3.9 Where  is a K-Vector of non-stationary endogenous variables that are commonly integrated of order one (I(1));  is a d-vector of exogenous deterministic variables;  ,  ,  and  are matrices of coefficients to be estimated and ∈ is a vector of innovation that may be contemporaneously correlated with their own lagged and all the variables on the right hand side.Due to the fact that many macroeconomic time-series are not stationary, the VAR model is generally estimated in its first difference form.Hence, the VAR model can be rewritten as follows: Where  = ∑  and r = − ∑ To determine the number of cointegrating vectors, Johansen (1988) and Johansen and Juselius (1990) proposed two statistical testes.These are the trace test and the maximum Eigen test.They are computed as follows: Where T is the number of usable observations or sample size and 's are the estimated Eigen values from the matrix.The trace test tests the null hypothesis of r cointegrating vectors against the alternative hypothesis of n cointegrating vectors, while the max Eigen test tests the null hypothesis of r cointegrating vectors against the alternative hypothesis of r+1 cointegrating vectors.
The Error Correction Model (ECM) was used to estimate the short-run behaviour of the variables of the study.Specifically, the ECM was used to measure the speed with which any disequilibrium in the short-run is adjusted to long-run equilibrium trend.Equation 3.6 is therefore expressed in ECM form as follows: Where  is the constant term or drift parameter, ∆ is the first difference operator, the terms with the summation sign (∑) (i.e. −  ) are the short-run coefficients, p is the ECM lag length, log is the natural logarithm,  is the error correction term coefficient, ECT is the error correction term, and ∈ is the white noise error term.All other variables are as earlier defined.
The Granger causality test was used to determine the nature and direction of causality between the variables.

Result and Discussion of Findings Descriptive Statistics
The results of the descriptive statistics are presented in Source: Computed from E-view Note: *denotes rejection of the null hypothesis of unit root at the 1 per cent level of significance.
From the ADF unit root test result in table 2, none of the variables are stationary at levels.However, they all become stationary at first difference (i.e, I(1)) at the 1 per cent level of significance.

Cointegration Test Result
The result of the Johansen cointegration test is presented in table 3. The trace test and max-Eigen statistic are the standard test statistic used in evaluating the Johansen cointegration test result.From the error correction model result in table 5, the coefficient of the error correction term (i.e.ECM (-1)) has a correct negative sign.It is also statistically significant at the 0.05 level of significance.The coefficient of the error correction term is -0.177055.This means that any disequilibrium in the short run is reconciled to long run stable equilibrium trend with a speed of adjustment of about 17% within a year.

Granger Causality Test Result
The result of the Pair-wise Granger causality test is presented in table 6

Long-run Regression Result
The estimated long-run result showed that foreign remittances inflows, government expenditure on agriculture and loans guaranteed under the Agricultural Credit Guarantee Scheme Fund (ACGSF) have significant positive impact on agricultural output.Bank credit to agriculture has insignificant positive impact on agricultural output, while exchange rate has insignificant negative impact on agricultural sector output in Nigeria.The economic implication of this result is that an increase in foreign remittances inflows, an increase in government expenditure on agriculture, an increase in loans guaranteed under the ACGSF will contribute to improvement in agricultural output performance in Nigeria.On the hand, an increase in exchange rate has adverse effects on agricultural sector output performance.

Short-Run Regression Result
The parsimonious error correction model result indicated that foreign remittances inflows in the current period and its lagged value in period one have insignificant negative impacts on agricultural output.However, the lagged value of foreign remittances inflows in period 2 has insignificant positive impact on agricultural output.Government expenditure on agriculture in the current period has insignificant negative impact on agricultural output.However, the lagged values of government expenditure in periods one and 2 have insignificant positive and significant positive impact respectively on agricultural output.The Agricultural Credit Guarantee Scheme Fund in the current period has significant positive impact on agricultural output.Bank credit to agriculture in the current period and its lagged value in periods 1 and 2 have insignificant positive impact on agricultural output.Exchange rate in the current period and its value lagged by 2 periods have insignificant positive impact on agricultural output.
The coefficient of determination (R-squared) is 0.769992.This means that the explanatory variables jointly account for about 76 per cent of the total variations in agricultural output.The Adjusted R-squared measures the change in the R-squared resulting from the loss of degree of freedom when additional explanatory variables are included in the model.It therefore measures the penalty for including irrelevant explanatory variables in the model.The adjusted R-squared is 0.565540.The Fstatistic is 3.766124 with a probability value of 0.00417.This shows that the overall estimated regression model is statistically significant at the 0.05 level of significance.The Durbin-Watson statistic is 2.865175 which imply that the estimated model is not affected by the problem of autocorrelation.
The Granger causality test result indicated only one unidirectional causality running from agricultural output to foreign remittances inflows.
The findings from this study on the impact of foreign remittances on agricultural sector performance inflows support the findings of Adejumo and Ikhide (2017) and Akpan et. al (2017).Similarly, the findings on the impact of government expenditure on agriculture support the findings of Okoh (2015).

Conclusion and Recommendations
The study finding indicates that foreign remittances inflows make strong contribution to output performance of the agricultural sector in Nigeria.Also government expenditure on agriculture significantly stimulates the growth of agricultural output in Nigeria.In addition credit allocated by the deposit money banks to the agricultural sector makes insignificant contribution to the output performance of the sector in Nigeria.Further, the amount of loans guaranteed under the Agricultural Credit Guarantee Scheme Fund of the Central Bank of Nigeria makes significant positive contribution to agricultural sector performance in Nigeria.Lastly, exchange rate reduces the output performance of the agricultural sector in Nigeria in an insignificant manner.Based on the conclusions drawn from the study, the following policy measures are recommended i.There is the need to encourage the recipients of foreign remittances to invest in the agricultural sector.To achieve this, the government should establish an institutional platform through which recipients of foreign remittances will be trained and enlightened on the investment opportunities in the agricultural sector where they can invest what they receive.In addition, migrants should be also be encouraged to invest directly in the country's agricultural sector.ii.The deposit money banks should be persuaded to allocate more credit facilities to productive ventures in the agricultural sector.iii.The Central Bank of Nigeria should increase the amount of loans guaranteed under its Agricultural Credit Guarantee Scheme Found.The CBN should also monitor beneficiaries of its ACGSF to ensure that such loans are actually invested in productive enterprises in the agricultural sector.iv.The CBN should also adopt a managed floating exchange rate regime so as to reduce the negative effects of high exchange rates on the performance of the agricultural sector.

table 1 Table 1 :
Descriptive Statistic Results Source: Computed from E-viewFrom the descriptive statistics results in table 1, the mean values of the variables are409.3510,0.467000,6.475750, 36.43075,0.143775,0.143775, and 10.04825 for AGOUT, REMT, GEXA, BCA, ACGSF, and EXR respectively.The standard deviation shows that AGOUT (464.6785) is the most unstable variable, while ACGSF (1.791479) is the least unstable variable.The skewness shows that all the variables are positively skewed.The kurtosis statistic shows that all the variables are leptokurtic (i.e., their values are greater than 3).This means that they have heavier or flatter tails relative to normal distribution.It also suggests that the variables are non-stationary.Hence, there is the need for stationarity test.

Table 2 :
ADF Unit Root Test Result

Table 3
Johansen Cointegration Test ResultFrom the Johansen cointegration test result in table 3, the trace test indicated 2 cointegrating equations, while the Max-Eigen test indicated 1 cointegration equation.The result therefore shows the presence of long-run equilibrium relationships between the explanatory variables and dependent variableLong Run Regression resultThe long-run coefficients obtained from the normalized cointegrating coefficients are presented in table 4.

Table 4 :
Long-Run CoefficientsThe figures in the first and second parentheses are the standard errors and t values respectively.

Table 6 :
Granger Causality Test Result

Table 7 :
Post Estimation Tests Results