Effect of Land Allocation on Agricultural Productivity in Nigeria Using ARDL Approach to Cointegration

Abstract

To test the validity of the inverse relationship between land productivity and land size within the context of Nigeria, this study employed the use of Autoregressive Distributed Lag (ARDL) model to test for the presence of long-run relationship between the two variables using data from 1980 to 2017 sourced from the Food and Agriculture Organisation. Prior to the test for cointegration, the Modified Granger Causality of the Toda and Yamamoto (1995) was used to test for the causality to be sure that causality indeed runs from land size to land productivity. The Granger Causality confirmed that land size Granger-cause land productivity. Given this result, the semi-log ARDL model was estimated to test for cointegration and also estimate the elasticity of the relationship in both the short and long runs. The result showed that while land size has inverse relationship with land productivity in both the short and long run, its influence in determining yield was only significant in the short run. The Error Correction Term, ECT(−1) which is the speed of adjustment of short-run disequilibrium to long-run steady state was both negative and significant, implying that there is going to be convergence to long-run equilibrium at the speed of about 24%.

Keywords

ARDL land productivity land size maize Nigeria

Introduction

Paradoxically, understanding the inverse relationship between land size and land productivity creates the exact mirror of the popular economics of scale. Economics of scale holds that although the cost of production may rise, the average cost of production is expected to decline, Rodríguez-Villalobos et al. (2018). On the contrary, the inverse relationship of land size and land productivity creates an opinion over the years among scholars that as land size increases, land productivity declines. This technically defeats the concept of commercial agriculture which has been the tool propagated by many policy makers as the way out of the food insecurity situation in developing nations and also a way to improve the economies of these nations by reducing their food import bills. Land is a natural resource in fixed quantity; the earlier the volume of production can increase with the size the better. Allocation of land in the midst of growing population in Nigeria that will require land for building homes, industrial and manufacturing revolution for economic development creates policy issues. This is because the same growing population will require food to survive.

Land is the most important factor of production without which no production can occur in most if not all of the sectors of the economy including agricultural sector. Without land, all other factors of production such as capital, labour and management will be idle and unproductive. Regardless of the exquisite combination of all other factors, even in their most optimal state, without the land’s natural ability whether in quality and texture for construction of factories and in soil nutrients for growth of crops and in size availability, expected yield and/or performance of the business is in danger. Productivity of land depends on its management and administration, Ladvenicová and Miklovičová (2015). Land as a factor of production is key to agricultural sector because it is the resource on which all other inputs are combined to achieve output. Land as a productive resource exhibits linear relationship with factors that determines its productivity. Such factors include use of improved seeds and/or seedling, use of fertiliser and herbicides, adequate agricultural information and education, credit facilities, machineries, and so on.

The inverse relationship between land size and land productivity has been an age-long discussion, Ananda Vadivelu et al. (2001), with some studies (e. g., Ali & Deininger, 2014; Assunção & Braido, 2007; Barrett et al., 2010; Carletto et al., 2013; Foster & Rosenzweig, 2017) agreeing that the inverse relationship was true while some other studies (e. g., Desiere & Jolliffe, 2018; Gollin & Udry, 2019; Gourlay et al., 2017) have noted that the inverse relationship could be due to measurement error especially as regards to other market and non-market factors that could influence land productivity, even soil variables. Some studies (Assunção & Braido, 2007; Barrett et al., 2010), however, have also attempted to check other factors to which the inverse relationship could be attributed but still found that the inverse relationship seems to be far from a statistical artifact that could be linked to measurement error, Carletto et al. (2013). According to Rahman and Rahman (2009) technology has a role to play in the relationship between land size and its productivity arguing that the relationship was positive in countries with advanced technologies and the inverse relationship existed in the developing nations.

Assunção and Braido (2007) employed plot-level data to test for the inverse relationship between land size and land productivity, attempting to know if cross-sectional heterogeneity could be responsible for the inverse relationship. They found out that the inverse relationship held true across different plot sizes used for the study, concluding that the puzzle of the inverse relationship between the two variables cannot be explained through cross-sectional heterogeneity. According to Masterson (2007), land quality difference may also be responsible for the inverse relationship. To this end, Barrett et al. (2010) added precise soil quality measurement at the plot level and factor market imperfections to check if these factors would influence the relationship. They found out very small portion of the inverse relationship could be explained by the factor market imperfections and none could be attributed to omission of soil variables. Hence, it is possible to model the relationship in a bivariate model. Carletto et al. (2013) concluded that using an improved measurement for land size only further agreed with the inverse relationship hypothesis.

In Uganda, Noack and Larsen (2019) in their study which used a large household level panel data found a negative relationship between land size and land productivity which was measured in output per unit of land measured in revenue terms. Although, Ren et al. (2019) found out that farm net profit increase with increase in farm size, they submitted that they are not able to conclude on the relationship between farmland size and overall productivity and therefore recommended the relationship for further research. Ali and Deininger (2014) attempted to check whether the hypothesis of inverse relationship between land productivity and land size was true at a global level. They revisited the hypothesis using plot-level data within the context of Rwanda, a country where small farm sizes were considered as hindrances to agricultural growth. According to their finding, there was a strong inverse relationship between land productivity and land size.

Wickramaarachchi and Weerahewa (2018) in their study on Paddy Rice Farmers in Irrigation Settlement in Sri Lanka found out that farmers with land size below 1.5 acres had lesser yield per acre measured in kg/acre compared to farmers with land size of at least 1.5 acres. While the yield of farmers with farmland below 1.5 acres was estimated to be about 1,733 kg/acre, the yield of farmers with at least 1.5 acres of farmland was estimated at about 1,921 kg/acre. The study also found a positive relationship between land size and land productivity stating that a 1% increase in farm size is likely to yield about 75 kg/acre of paddy rice. In Rwanda, Mathias (2010) studied how land fragmentation affects productivity and technical efficiency of smallholder maize farmers. While the study noted that average land area available to the maize farmers was diminishing over the year, from 1.2 ha in 1984 to 0.84 ha in 2002 and to 0.72 ha in 2006, households with at least two plots of land were about half of the sampled farmers in the Southern region of Rwanda. They study found out that smallholder maize farmers were technically efficient.

Kiani (2008) established a relationship between farm size and productivity in Pakistan using Cobb-Douglas production function and found out that productivity is different with farm size, noting that it is high among the smallholder farmers due to the compactness for labour engagement and irrigation system, it is low among the medium scale farmers due to inefficient input combination and also high among the large-scale farmers because of maximum capital investment being used for farm operations. Vigneri (2008) also found out that smallholder cocoa farmers had higher productivity. Kimhi (2003) assessed the relationship between plot size and productivity among Maize farmers in Zambia using two stage least square (2SLS) and two-sided Tobit model. The study found out that there is a positive relationship between plot size and maize production. Using both parametric and non-parametric analyses, Masterson (2007) found out that there is inverse relationship between farm size and productivity in Paraguay and also said net farm income per hectare is higher among the smallholder farmers.

Sheng et al. (2019) used farm-level panel data of maize farmers in China from 2003 to 2013 to investigate the inverse relationship of land productivity and land size. Having controlled for cross-sectional heterogeneity, they still found the U-shape curve which implies that the inverse relationship is valid and that is exist among the smallholder farmers. Thapa (2007) in the study of Nepal farmers found out that small farms were more productive compared to the large farm due to their ability to intensively use the little productive resources such as cash and labour available to them. Attempting to understand the role tenure security plays in the relationship between land size and its productivity, Tatwangire and Holden (2013) used panel data that relied mostly on global positioning system (GPS)-measured plots in order to minimise measurement errors to test for freehold, Mailo tenure system and customary system and found out that the inverse relationship existed in the three tenure systems considered.

Ansoms et al. (2008) in their own study in Rwanda although found the inverse relationship between land size and land productivity to be true, but stressed the need not to see this from the perspective of efficiency alone, noting that land scarcity can force farmers to overexploit their farmlands. Obasi (2007) in the case of Nigeria concluded that lack of quality inputs among the smallholder farmers is responsible for the inverse relationship between land size and land productivity. Chen et al. (2011) could not find the inverse hypothesis to be true in the case of China. According to Tamel (2011), there are mixed finding in the case of the United States of America in the bid to ascertain the validity of the inverse relationship. The study revealed that in some areas, the inverse relationship holds and some other areas, the hypothesis did not hold.

Why Maize Enterprise?

In Nigeria, maize as a crop enterprise has not been given the due attention it deserves. It is seen more as a food crop than a cash crop that has huge market potential given its importance in the growth of livestock sector as one of the major ingredient of feed formulation for livestock animals. For a developing nation like Nigeria which is about 200 million in population and still growing, urbanisation will create increasing demands for livestock products, Food and Agriculture Organisation (FAO) (2018), such as frozen chicken, eggs and milk. Therefore, maize offers a multi-billion-naira market whose potentials when fully tapped into will offer gainful and profitable employment to the unemployed youth and ensure a sustained animal protein market and also guarantee sustainable living for the maize farmers. The demand for maize as a major raw material by feed mills in the country for poultry feeds, fish feeds and other livestock animals have been greatly increasing. From 1990 to 2018, maize consumption has been on the increase from 5,388 MT in 1990 to about 11,631. 30 MT except for 2000 to 2002 that maize consumed was below 5,000 MT, with a projected consumption of 13,787.83 MT by 2027, Organization for Economic Cooperation and Development (OECD)-FAO (2018). Nigeria as at 2018 consumed about 293 MT of poultry meat, about 393.88 MT of beef meat, 1,910.23 MT of fish and about 256.63 MT of dairy products. By implication, a performing maize enterprise implies better life for the farmers.

Generally, in Nigeria, there is a dearth of study on the relationship between land size and land productivity on one hand, and on the other hand, maize enterprise has not been largely attended to in terms of scientific studies. This study thus intended to not only examine the validity of the inverse hypothesis in Nigeria but also focused its attention on the enterprise that has been relegated or regarded as not so important. As noted above from the OECD-FAO data, maize which is regarded as food crop is fast becoming a cash crop with the very crucial roles it plays in the livestock industry. For Nigeria, being an agrarian nation that is largely smallholder driven, it is important to test the hypothesis to check for its validity and hence to suggest policy drives to enhance higher yield for the land, given the economic importance of maize. Therefore, the research questions this study attempted to provide answers to include:

What is the causal relationship between land size and land productivity in Nigeria?

Is there the presence of cointegration between the productivity of land and its size in Nigeria?

How do land productivity and land size interact to determine farmers’ yield in the short run in Nigeria?

The general objective of this study was to test for the validity of the inverse relationship between land size and land productivity in Nigeria. The specific objectives of this study were:

To examine the causality between land size and land productivity

To test for the presence of cointegration between land size and land productivity and

To estimate the short-run relationship between land size and land productivity.

Hypothesis of the Study

To examine the validity of the inverse relationship between land size and land productivity, the relationship is tested given the following hypothesis:

H₀: that the relationship between land size and land productivity is not inverse

I n Y D = β_{0} + β_{i} LS

H₁: that that the relationship between land size and land productivity is inverse in nature

I n Y D = β_{0} - β_{i} LS

To extend the works done already in this regard, this study first employed the Granger Causality Test to check the causal relationship between the land productivity and land size to be sure that indeed, the causal relationship is running from land size as the exogenous variable to the land productivity as the endogenous variable.

Methodology

Study Area

Nigeria, a country in the West of the continent of Africa has a land size of 923,768 km² and a population 209,210,535 people (National Population Commission) to feed. The population is estimated to grow to as high as 400 million by 2050, with about 72% being urban population from the current 52% urban population.

Data

This study relied on the time series data from 1980 to 2017 sourced from the FAO Statistics (FAO-STAT). Yield of maize for the period was used to proxy productivity of maize farmers in Nigeria. Hence, the variables for this study are as stated in Tables 1 and 2.

Table 1.

Description of Variables.

Variable	Description
Dependent variable YD	Maize yield per hectare (tons/ha)
Independent variable InLS	>Natural logarithm of land area of harvested maize

Source: Author’s computation using Eviews Software version 10.

Table 2.

Statistical Description of Variables.

Item	YD	InLS
Mean	1.29 ton/ha	3,920.6 (‘000ha)
Median	1.27 tons/ha	3,864.5 (‘000ha)
Maximum	2.2 tons/ha	7,212.1 (‘000ha)
Minimum	0.57 tons/ha	438 (‘000ha)
Jarque–Bera Stat	0.3397	4.5430
Probability	0.8438	0.1032

Source: Author’s computation using Eviews Software version 10.

Analytical Tool

To examine the validity of the inverse relationship between land size and land productivity in the case of Nigeria, this study adopted the use of the Toda and Yamamoto (1995) Modified Granger Causality test due to the possibility of the existence of co-integration between land productivity and land size, a situation for which the normal Granger Causality may not be able to produce robust estimate. This study, therefore, extended its finding to investigate the presence of co-integration between land productivity and land size. To examine the presence of long-run relationship between the two variables, the Autoregressive Distributed Lag (ARDL) Model was adopted. Also, the semi-log functional form of the regression model was used due to the statistical nature of the data; the land size variable was not normally distributed and hence was transformed into its natural logarithm.

Results and Discussion

Because agriculture is seasonal and can easily be influenced by changes in climatic variables such as precipitation and temperature, the result of the ADF unit root test with both intercept and trend is presented in Table 3.

Table 3.

Granger Causality Test

The Toda and Yamomoto (1995) Modified Granger Causality test function that was examined to investigate the inverse relationship between land productivity and land size in Nigeria was as stated as:

I n Y D = β_{0} + \sum_{i = 1}^{k} b_{1} {InYD}_{t- 1} + \sum_{j = k + 1}^{d_{m a x}} b_{2} {InYD}_{t- 1} + \sum_{i = 1}^{k} γ_{1} I n L S_{t - 1} + \sum_{j = k + 1}^{d_{mat}} γ_{2} I n L S_{t - 1} + ϵ_{t}

(1)

The alternative function to the above granger-causality inverse model is stated as:

I n L S = α_{0} + \sum_{i = 1}^{k} α_{1} I n L S_{t - 1} + \sum_{j = k + 1}^{d_{max}} α_{2} I n L S_{t - 1} + \sum_{i = 1}^{k} δ_{1} I n Y D_{t - 1} + \sum_{j = k + 1}^{d_{max}} δ_{2} I n Y D_{t - 1} + ϵ_{t}

(2)

Hence, the hypothesis under investigation is given as:

H₀: Land size does not granger cause land productivity

H₁: Land size granger cause land productivity.

With optimal lag k = 1 and d_max = 1, (k + d_max) Vector autoregressive for YD and InLS along with the second lags of the variables was estimated. The Granger Causality test result is stated in Table 4.

Table 4.

Toda and Yamamoto Granger Causality Test Result.

Independent Variable	Dependent Variable: InLP			Dependent Variable: InLS
Independent Variable	X²	df	P Value	X²	df	P Value
InLS	7.764	2	.0206**	–	–	–
YD	–	–	–	1.631	2	.4424

Source: Author’s computation via Eviews 10.

Note: Significant at 5%

The Granger Causality test shows that the null hypothesis that land size does not granger-cause land productivity cannot be accepted. Hence, in the case of Nigeria land size has influence on the productivity of the producers. Having established the functional relationship between land productivity and land size in Nigeria, to test for the inverse relationship, the ARDL Model was applied. However, to establish the order of integration of the variables, the ADF unit root test for each of the variable was estimated using the function:

Δ Y_{t} = \propto + ρ Y_{t - 1} + \sum_{i - 1}^{j} Y Δ Y_{t - j} + U_{t}

(3)

Bounds Test Approach to Co-integration

With the functional relationship of land productivity and land size in Nigeria established to be $InLP = f (InLS)$ , the Pesaran and Shin (1999) and Pesaran et al. (2001) ARDL model functional form was given as:

Δ I n L P = α_{0} + \sum_{i = 1}^{n} \emptyset_{1} {ΔInLP}_{t-i} + \sum_{i = 0}^{n} \emptyset_{2} Δ I n L S S_{t - i} + \partial_{1} I n L S_{t - 1} + \partial_{2} I n L P_{t - 1} + ϵ_{i}

(4)

where ∆ = first difference operator and is the error term.

The F-statistic was used to test for the null hypothesis of no presence of co-integration. According to Okunola (2017), if the F-statistic is below the lower bound I(0) critical value, the null hypothesis of no presence of co-integration between land size and land productivity cannot be rejected and if it is above the upper bound I(1) critical value, the null hypothesis of no co-integration between the economic variables cannot be accepted. However, if the F-statistic falls between the lower and upper bounds critical values, the result is deemed inconclusive. The F-statistic was compared to the critical values according to Narayan (2005) due to the small sample size nature of this study. This is based on the fact that the Pesaran et al. (2001) critical values are based on large sample size. The result of the cointegration test which was based on optimal lag 1 and Schwarz Information Criteria model selection method is stated in Table 5.

Table 5.

Bounds Test for Cointegration.

CV (Restricted Intercept and No Trend)	Lower Bound	Upper Bound
1%	5.593	6.333
5%	3.937	4.523
10%	3.21	3.73

Source: Author’s computation via Eviews 10.

Notes: Calculated F-statistics = 4. 04 at k = 1. CC, critical value.

Selected model: ARDL(1,1).

From the estimates, the F- stat is greater than the critical value at 10%, hence, the conclusion to reject the null hypothesis of no presence of long-run relationship between land productivity and land size among maize farmers in Nigeria, although the long-run relationship is a very weak one. The implication of this is that the size of land is seemingly not the only factor determining whether maize farming in Nigeria is going to be productive or not. While the smaller the land size, the lower the farm-level investment that is required to improve on the performance of the land by consuming and/or adopting soil conservation techniques, these technologies such as use of improved fertiliser, improved maize varietiesherbicides and even irrigation system all combine with the size of farmland to give the farmers the expected productivity

Long-run Elasticities

With the presence of long-run relationship between land size and productivity, the long-run elasticities based on Equation (5) is estimated.

YD = \propto_{0} + \partial_{1} I n L S_{t - 1} + \partial_{2} {YD}_{t - 1} + ϵ_{i}

(5)

The estimates of the elasticities are given in Table 6.

Table 6.

Long-run Elasticities.

ARDL(1,1) Selected Based on Schwarz Information Criterion
Variable	Coefficient	Standard Error	T-statistics	P Value
InLS	−0.3545	0.4053	−0.8744	.3880
Constant	2.8964	1.4860	1.9491	.0596*

Source: Author’s computation via EViews 10.

Note: *Significant at 10%.

As noted from the cointegration test that the long-run relationship between land productivity and land size is a weak one significant at 10%, the long-run elasticity has also reinforced this fact given that it is not significantly influencing land productivity in the long run and in fact has a negative relationship. This implies that as Nigeria progresses into the future with a rapidly growing population demanding for land areas for non-agricultural activities and due to the human activities increasing the rate at which the climate is changing, land size is not likely going to be a significant factor determining land productivity. This is in part valid because consistent use of the land without appropriate soil conservation techniques will keep diminishing the performance of the land. Also, increasing temperature and almost not enough precipitation due to consistent increase of the CO₂ content in the atmosphere will continue to cause damages to the development of maize seeds during their growth stage. Naturally, maize requires a temperature between 21˚C and 27˚C in the warmest period, Kamara et al. (2020). Regardless of the land size, if other factors such as irrigation system, planting of heat-resistant maize seed varieties or drought resistant maize seed varieties are not in place, the productivity of land may not justify the efforts invested into production activities.

Short-run Elasticities

The short-run elasticities and the error correction term (ECT) of the model was estimated using Equation (6):

Δ I n L P = \propto_{0} + \sum_{i = 1}^{n} \emptyset_{1} Δ I n L P P_{t - i} + \sum_{i = 0}^{n} \emptyset_{2} Δ I n L S_{t - i} + ϵ_{i}

(6)

The result of the analysis is presented in Table 7.

Table 7.

Short-run Estimated Parameters.

Variable	Coefficient	Standard Error	T-statistics	P Value
∆InLS	−1.2758	0.938	−4.423	.001***
ECT	0.370	0.661	−3.823	.011***

Source: Author’s computation via EViews 10.

Note: ***Significant at 1%.

The short-run elasticity shows that land size is significantly influencing land productivity in the short run. In order words, because land is still available in appreciable sizes to maize farmers, it is playing significant role in determining the productivity of land. Connecting this estimate to the long-run relationship it points to a dangerous future for the most populous African country that has a young and highly productive population given that as the country moves into the future demand for land will become competitive for both agricultural and non-agricultural uses. The elasticity depicted from the short-run estimates implies that for every 1% increase in land size, there will be 0. 013% unit (tons/ha) change in land yield. This further points to the fact that acquiring more land does not translate to higher yield even though land size significantly influence yield. Hence, with researches into improved varieties of maize that can withstand both drought and heat, and construction of irrigation systems, it is left for the farmers to adopt these improved varieties and other measures such as climatic resilience measures to increase their yield per hectare cultivated. However, there is also the need for policies and legislative framework that can conserve land for agricultural purposes for the country to be able to make maize available for its teeming population that would require maize for both human consumption and industrial needs especially the livestock feed industry. The short-run relationship between land productivity and land size is an inverse one which implies that as land sizes increases, productivity in terms of yield declines. While this may be subject to debates of economics of scale, some studies (e. g., Kiani, 2008) have noted the larger the farm size, the more it becomes difficult to manage the resources, combine them to achieve optimal output. The finding of this study agrees with finding of (Desiere & Jolliffe, 2018; Gollin & Udry, 2019; Gourlay et al., 2017) where all affirmed the inverse relationship between land productivity and land size.

Residual Diagnostics Tests

To ascertain the validity of the model for economic interpretation, the diagnostics tests for the residuals were carried out. On the basis of the results of the tests, the model estimated for this study is normally distributed, there is no presence of serial correlation and it is homoscedastic. The results of the test are as stated in Table 8.

Table 8.

Residual Diagnostic Test.

Test for Normality
Jarque–Bera	0.435	Probability (Jarque–Bera)	.785
Breusch–Godfrey Serial Correlation LM Test
Obs*R-squared	1.038	Probability Chi-square (5)	.485
Heteroskedasticity Test: White
Obs*R-squared	3.677	Probability Chi-square (27)	.813

Source: Author’s computation via EViews 10.

Note: LM, lagrange multiplier.

Stability Test of the Model

As recommended by Brown et al. (1975), the CUSUM and CUSUMSQ tests were carried to ascertain the stability of the model. The residuals rests indicated stability in the coefficients of the parameters. While the CUSUM test was applied to the sum of recursive residuals, the CUSUMSQ was applied to the squared residuals. If the cumulative sum plot goes outside the 5% critical lines area, the coefficients of the parameter estimates cannot be said to be stable. The results are presented in Figures 1 and 2.

Figure 1.

Source: Author’s computation using Eviews Software version 10.

Figure 2.

Source: Author’s computation using Eviews Software version 10.

Summary, Conclusion and Recommendation

Land is one of the major factors of production if not the most critical. It is on land that all other factors of production such as capital, labour and management skills are combined to achieve optimal productivity. However, there are other inherent factors such as soil nutrients and compositions as well as weather conditions that must be fulfilled for different crops to grow. Over time in the literature, there has been the empirical argument that land size has a negative relationship with land productivity. In other words, as land size increases, land productivity tends to decline. While some researchers support this notion, some researchers have found reasons to disagree with it. Although, Nigeria as an agrarian nation and also a growing nation needs to give attention to land apportionment for agricultural production and in fact protection of farming villages and settlements to ensure that there is land for food production. However, more than making lands available, the productivity of the land should be paramount and other resources such as credit support, quality extension services and market information should be made available to the farmers for the to be profitable. A large land without commensurate investment in the land for productivity will not yield so much for the farmer. These farm-level investments include consumption and/or adoption of soil conservation techniques, improved inputs as well as having access to production information such as weather forecast and successful researches relevant to maize production.

In order to ascertain the validity of the inverse relationship between land productivity and land size in Nigeria, the Modified Granger Causality test was carried out to check for the causality and direction of causality between the two variables. Having established the causal relationship between land size and land productivity, with land size as the independent variable, ARDL model was used to analyse the data from 1980 to 2017 sourced from the Food and Agriculture Organisation Statistics, FAO-STAT. Before checking for the presence of cointegration, the ADF unit root test was carried out to be sure of the order of integration of the variables. The unit root test showed that both variables are of I(1) order of integration. The ARDL confirmed the presence of cointegration between land productivity and land size, albeit a negative, weak and insignificant long-run relationship. The short-run elasticities were estimated alongside the ECT which is the speed of adjustment of the model from a short-run disequilibrium to a long-run stability. Both the land size and land productivity exhibited negative relationship in the context of Nigeria, however the influence land size has in determining the land yield in the short run was highly significant.

Now that this study has been able to establish that there is a causality running from land size to land productivity, thus implying the smaller the size of the land, the better the farmer is able to manage and monitor is performance and the lower the cost required to cultivate such lands. As much as this study has been able to establish this hypothesis, it has also given rise to other questions such as how access to agricultural credit as an instrumental variable can influence the inverse relationship between land size and land productivity. One of the major challenges of increasing size of land being used for agricultural activities will be the finance to maximise the land asset. Also, how environmental factors such as precipitation, temperature and humidity and generally the drivers of climate change will affect this inverse relationship is important. The security of land tenure system also can play a role in the optimisation of land as a factor of production by the farmer. No farmer will invest in a land whose ownership can change in the nearest future. All of these will enable the policy makers to avoid all-size-fit-all policies in the bid to develop the agricultural sector in Nigeria.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author received no financial support for the research, authorship and/or publication of this article

ORCID iD

Akinbode Okunola

References

Ali

D.A.

, & Deininger

(2014). Is there a farm-size productivity relationship in African Agriculture? Evidence from Rwanda [Policy Research Working Paper No. 6770]. World Bank.

Ananda Vadivelu

, Wani

S.P.

, Bhole

L.M.

, Pathak

, & Pande

A.B.

(2001). An empirical analysis of the relationship between land size, ownership, and soybean productivity: New evidence from the semi-arid tropical region in Madhya Pradesh, India [Natural Resource Management Program Report No. 4] (pp. 50). International Crops Research Institute for the Semi-Arid Tropics. https://doi.org/10.3368/le.91.2.317

Ansoms

, Verdoodt

, & Van Ranst

(2008). The inverse relationship between farm size and productivity in rural Rwanda [Discussion Paper/2008. 09]. Institute of Development Policy and Management, University of Antwerp.

Assunção

, & Braido

(2007). Testing household-specific explanations for the inverse productivity relationship. American Journal of Agricultural Economics , 89 4, 980–990. https://doi.org/10.1111/j.1467-8276.2007.01032.x

Barrett

C.B.

, Bellemare

M.F.

, & Hou

J.Y.

(2010). Reconsidering conventional explanations of the inverse productivity–size relationship. World Development , 38, 88–97. https://doi.org/10.1016/j.worlddev.2009.06.002

Carletto

, Savastano

, & Zezza

(2013). Fact or artifact: The impact of measurement errors on the farm size–productivity relationship. Journal of Development Economics , 103, 254–261. https://doi.org/10.1016/j.jdeveco.2013.03.004

Chen

, Huffman

W.E.

, & Rozelle

(2011). Inverse relationship between productivity and farm size: The case of China. Contemporary Economic Policy , 29, 580–592. https://doi.org/10.1111/j.1465-7287.2010.00236.x

Desiere

, & Jolliffe

(2018). Land productivity and plot size: Is measurement error driving the inverse relationship? Journal of Development Economics , 130, 84–98. https://doi.org/10.1016/j.jdeveco.2017.10.002

Foster

A.D.

, & Rosenzweig

(2017). Are there too many farms in the world? Labormarket transaction costs, machine capacities and optimal farm size [NBER Working Paper] (Vol. 23909). https://doi.org/10.3386/w23909

10.

Gollin

, & Udry

(2019). Heterogeneity, measurement error and misallocation: Evidence from African agriculture [NBER Working Paper] (Vol. 25440). https://doi.org/10.3386/w25440

11.

Gourlay

, Kilic

, & Lobell

(2017). Could the debate be over? Errors in farmerreported production and their implications for the inverse scale-productivity relationship in Uganda [Policy Research Working Papers] (Vol. 8192). World Bank. https://ssrn.com/abstract=3036194

12.

Kiani

A.K.

(2008). Farm size and productivity in Pakistan. European Journal of Social Sciences , 7 2, 42–52.

13.

Kimhi

(2003). Plot size and maize productivity in Zambia: The inverse relationship re-examined . Centre for Agricultural Economic Research and Rural Development Research Consortium, The Hebrew University of Jerusalem. https://doi.org/10.22004/ag.econ.14980

14.

Ladvenicová

, & Miklovičová

(2015). The relationship between farm size and productivity in Slovakia. Visegrad Journal on Bioeconomy and Sustainable Development , 4 2, 46–50. https://doi.org/10.1515/vjbsd-2015-0011

15.

Masterson

(2007). Productivity, technical efficiency, and farm size in Paraguayan agriculture [Working Paper No. 490]. Levy Economics Institute of Bard College. https://dx-doi-org-s.web.bisu.edu.cn/10.2139/ssrn.964722

16.

Mathias

(2010). The effect of land fragmentation on the productivity and technical efficiency of smallholder maize farms in southern Rwanda [Research theses 243457]. Collaborative Masters Program in Agricultural and Applied Economics.

17.

Noack

, & Larsen

(2019). The contrasting effects of farm size on farm incomes and food production. Environmental Research Letters , 14, 084024. https://doi.org/10.1088/1748-9326/ab2dbf

18.

Obasi

P.C.

(2007). Farm size–productivity relationships among arable crops farmers in Imo State, Nigeria. International Journal of Agriculture and Rural Development , 9 1, 91–99. https://doi.org/10.4314/ijard.v9i1.2673

19.

Okunola

A.M.

(2017). Long and short run dynamics of agricultural and petroleum sectors in the economic growth of Nigeria. Romanian Economic Journal , 20(64).

20.

Rahman

, & Rahman

(2009). Impact of land fragmentation and resource ownership on productivity and efficiency: The case of rice producers in Bangladesh. Land Use Policy , 26, 95–103. https://doi.org/10.1016/j.landusepol.2008.01.003

21.

Ren

, Liu

, Van Grinsven

, Reis

, Jin

, Liu

, & Gu

(2019). The impact of farm size on agricultural sustainability. Journal of Cleaner Production , 220, 357–367. https://doi.org/10.1016/j.jclepro.2019.02.151

22.

Rodríguez-Villalobos

, García-Martínez

, & Mata-Camarena

R.O.

(2018). Economies of scale and minimization of the cost: Evidence from a Manufacturing Company. Journal of Eastern Europe Research in Business and Economics , 2018(2018), 128823. https://doi.org/10.5171/2018.128823

23.

Sheng

, Ding

, & Huang

(2019). The relationship between farm size and productivity in agriculture: Evidence from maize production in Northern China. American Journal of Agricultural Economics , 101 3, 790–806. https://dx-doi-org.web.bisu.edu.cn/10.1093/ajae/aay104

24.

Tatwangire

, & Holden

S.T.

(2013). Land market participation and farm size–productivity relationship: Implications of land tenure reforms in Uganda. In Holden

, Otsuka

, & Deininger

(Eds), Land tenure reforms in Asia and Africa: Impacts on poverty and natural resource management (pp. 187–210). Palgrave Macmillan.

25.

Thapa

(2007). The relationship between farm size and productivity: Empirical evidence from the Nepalese mid-hills . 106th Seminar, October 25–27, 2007, Montpellier, France. European Association of Agricultural Economists. https://econpapers.repec.org/paper/agseaa106/7940.htm

26.

Toda

H.Y.

, & Yamamoto

(1995). Statistical inference in vector autoregressions with possibly integrated processes. Journal of Econometrics , 66(1-2), 225–250. https://doi.org/10.1016/0304-4076(94)01616-8

27.

Vigneri

(2008). Drivers of change in Ghana’s Cocoa Sector [Ghana Strategy Support Programme Working Paper No. 13]. International Food Policy Research Institute. https://www.ifpri.org/publication/drivers-change-ghanas-cocoa-sector

28.

Wickramaarachchi

N.C.

, & Weerahewa

(2018). Relationship between land size and productivity: Empirical evidence from paddy farms in irrigation settlements of Sri Lanka. Sri Lanka Journal of Economic Research, Sri Lanka Forum of the University Economists , 5 2, 25–142.

ADF Statistics
Variable	Critical Values					Remark
Variable	Levels	First Difference	1%	5%	10%	Remark
YD	−2.32	−7.44***	−3.55	−2.91	−2.60	I(1)
InLS	−0.64	−7.09***	−3.55	−2.91	−2.60	I(1)

Effect of Land Allocation on Agricultural Productivity in Nigeria Using ARDL Approach to Cointegration

Abstract

Keywords

Introduction

Why Maize Enterprise?

Hypothesis of the Study

Methodology

Study Area

Data

Description of Variables.

Statistical Description of Variables.

Analytical Tool

Results and Discussion

Top 20 Most Active Countries Publishing Articles on CGRM, Based on Citations.

Granger Causality Test

Toda and Yamamoto Granger Causality Test Result.

Bounds Test Approach to Co-integration

Bounds Test for Cointegration.

Long-run Elasticities

Long-run Elasticities.

Short-run Elasticities

Short-run Estimated Parameters.

Residual Diagnostics Tests

Residual Diagnostic Test.

Stability Test of the Model

Summary, Conclusion and Recommendation

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

References