Food security continues to play a significant role in the global macroeconomic dynamics of many countries. For policymakers and governments in developing and particularly in low-income countries, food security offers even further challenges both from the perspective of achieving SDGs targets and the welfare perspective of the many poor households in the global economies as documented in the literature. For instance, COVID-19 has continued to pose serious economic [
19] and welfare challenges to households in Sub-Saharan Africa [
16,
23]. By implication, this is an alarming aspect for the prospect of sustainable development of food security in Nigeria. We examine first the findings from the recent concepts. The study finds that food security has extensively been associated with increasing food availability and access for a population. Thus, we argue that it is an essential means of achieving the SDGs for poverty and hunger reduction, particularly when focused on small farmers. The physical, political, economic, and social settings, in which a family lives, however, have influenced the riskiness of specific social unrest. For example, rural economies in agrarian regions in Nigeria are more prone to shocks because of conflicts and overall insecurity and this has been imposing a fastening restraint on the long-run growth of agricultural production and domestic income ever since the emergence of unprecedented insecurity in Nigeria [
18]. Agriculture is vital to rural economies, and improvement in agricultural productivity can result in economic development or poverty alleviation because livelihoods are so important to food security and sustainability. However, this will only work in settings where people have access to their land. How well these people adapt to a certain threat is largely determined by the strength and diversity of their livelihood. The survival of the citizens largely depends on agricultural sources of income but social unrest, terrorism, and economic fluctuations have challenged food security sustainability in Nigeria [
18].
The findings from our results make some sense at least because they are in line with the recent concepts of the National Bureau of Statistics, FIES, FEWS, OECD, and World Bank in terms of underproduction challenges and food emergencies in Nigeria, e.g., see
Figure 1,
Figure 2,
Figure 3,
Figure 4,
Figure 5,
Figure 6 and
Figure 7. For instance,
Figure 1 exhibits the prevalence of food insecurity in Nigeria between 2014 and 2019, which increased with population growth in the country over time, where demand for food rose to 9.1 percent between 2017–2019.
Figure 2 demonstrates FIES analysis that food insecurity is more prevalent in southern Nigeria on account that northern Nigeria has been the food basket of Nigeria, but in
Figure 3, FEWS outcome countered the FIES analysis because outside of northern Nigeria, in much of the rest of the country, the agricultural season is progressing favorably with average and above-average harvests. Therefore, households in these areas will have access to food and income and will remain in minimal acute food insecurity while inside the north and areas affected by crisis and conflicts between farmers and pastoralists, among other social unrest are facing greater difficulty in accessing foods and income and will be stressed, famine or require emergencies,
Figure 4 illustrates that the incidence of moderate and severe food insecurity as per the FIES analysis rose in 2020 which justifies the FEWS analysis. Besides the findings from these institutions, there is also the existence of developing characteristics of Nigeria as an African country. In specific terms, it is argued that the system of agricultural food production alongside none mechanized farming operating in the country is constraining the ability of the agricultural sector is playing a vital role in the sustainable food security process in Nigeria [
7].
Next, we examine the empirical approaches. The discussion starts from
Figure 2, because it gave us a road map to proceed with the subsequent analyses. The findings revealed unique cointegration equations at a 5% (0.05) significance level, and the implication for this is, the null hypothesis of no cointegration is rejected. This shows the implication for the existence of long-run relationship among the
LAFVA,
LIDP,
LGDPpc, and
LPOPtl,
LREER, and
LINFcp up to the third null hypothesis for both Trace and the Max Eigen-value cointegration results. This result is in line with the findings of [
61]. The long-run relationship among
LAFVA,
LIDP,
LGDPpc,
and LPOPtl,
LREER, and
LINFcp can be explained as follows. First, the increase in value-added agricultural production and per capita GDP could positively enhance the goal for sustainable food security. Second, pressure from increasing internally displaced persons, exchange rate volatilities, population growth, and food inflation could ultimately affect the goal of achieving sustainable food security. For example, previous studies found that forced displacement and increase in population have an inverse relationship with agricultural food production [
18]. Similarly, food inflation reduces the availability and access to food, particularly at the local level. Thurs, our results are in line with the findings of [
17,
62], who find that increase in output and economic performance increase food security while exchange rate volatility, increase in population, and forces displacement are related to an increase in food insecurity in the long run. Next, we examine the ARDL Bounds test findings (e.g., see
Figure 3). The findings reveal that the calculated value of F- statistics of 6.38 is greater than the corresponding upper bound table value which is 3.99. The null hypothesis of no cointegration is rejected. Thus, the computed F-statistics 123.3378(0.0000) of the joint Wald test left us with a decision to retain all our variables in the model because the parameters associated with these variables are not zero. As shown by the F- statistics, the implication for this is, the
LAFVA,
LIDP,
LGDPpc, and
LPOPtl,
LREER, and
LINFcp are bound together because of equilibrium forces towards a long-run relationship [
57,
58]. These indicators provide information related to the food security or insecurity and hunger relation in the long run. For instance, a previous study has shown that food security and hunger may not always be linked, but they are related; if people are food insecure for months at a time, they may very well experience a substantial drop in food intake that leads to hunger. In general, agricultural system model analyses more commonly employ availability indicators but would provide improved guidance for research and programmatic efforts with a focus on indicators of food access [
8]. Next, we examine
Figure 4, the estimates of
LAFVA,
LIDP,
LGDPpc, and
LPOPtl,
LREER, and
LINFcp on the LFPI in the short and long run are statistically significant after the robustness checks, and diagnostic tests have been satisfied. The findings show that the model’s coefficient of ECMt-1 is negative (−0.0130 approximately) and statistically significant (0.0000) at 1 percent. As a result, the convergence of the disequilibrium to the long-run equilibrium requires, on average, 1.3% annually. This is a relatively sluggish adjustment, and it indicates that it will take more than 9 years on average for the long-run equilibrium to be fully restored. This result is supported by the United Nations SDGs’ target which requires developing countries to achieve at least 7% annual growth to meet the catch-up growth process gap between the developed and developing economies but the 7% target is still out of reach [
63,
64]. Going by these findings, there is high evidence to believe that accomplishing the SDGs target by the year 2030 for poverty and hunger reduction, particularly when intended for food security sustainability alongside small producers by the year, 2030 can be rarely achieved. This finding has not been investigated by the previous studies. Next, the findings from
Figure 5,
Figure 6 and
Figure 11, left us to retain all our variables because the model residuals are homoskedastic and serially uncorrelated. In addition, the functional form of the model is correctly specified, and its errors are normally distributed. On a final note, the coefficients of the model are also stable. To confirm the significance of the associated parameter for the explanatory variable in the model, the Wald test was employed, and the decision was to retain our variables. The study scrutinized the structural breaks by means of the cumulative sum of squares to ascertain the stability of the parameters in the estimated model, which is exhibited in
Figure 11, which follows Adedoyin et al. [
58], that the test statistic for the estimated parameters should be within the significant bounds of the 95% confidence interval and this was confirmed. Over time,
Figure 11, also confirmed the stability of the estimated coefficients as recommended, see Brown, Durbin, and Evans, 1975 [
60].