Agriculture is the main employment sector for the poor, employing 76.3% of the extreme and 60.7% of the moderate poor [1] . Most of this group tends to be subsistence or semi-subsistence oriented and faces significant barriers to entering higher value agricultural activities. The agrarian population in Cameroon is essentially small-scale peasant farmers and their family members, who make up about 70% of the agricultural population. In the face of an increasing population and settlement, the use of farm inputs that are not environmentally sustainable, and wide-scale agricultural and forestland use changes, food crop production trends seem to be uncertain or rather stagnant as studies report that projected/expected needed crop production is often actual production [2] . Argued that, between 1975 and 2005, there were 20 years during which actual cereal production in Cameroon was persistently below projected/expected needed cereal production level. Epule et al. (2012) verified the vulnerability of experiencing food shortages along gender and poverty lines. Many past studies have argued that, as in most sub-Saharan African countries, Cameroon is currently experiencing declines or stagnation in food production at the national scale, which could be attributed to its socio-economic characteristics [2] . Agricultural adaptation requires consideration of both human and physiographic challenges that are responsible in specific contexts [3] to verify the role of land management practices and the socio-cultural properties of small-scale farmers in establishing differences in crop yields. Access to food is a basic need for human beings; however, many poor people do not have physical and economic access to sufficient, safe, and nutritious food [4] . There are many factors that account for this condition of food security, such as loss of job, lower level of education and employment, lack of access to land, single-parent families, unstable income level, and having a poor family head. All these factors lead to poverty, and the fundamental outcome is inadequate access to food [5] .

In Cameroon, agriculture remains the backbone of the economy, employing up to 70% of the labor force and generating about 42% of gross domestic product (GDP) and 30% of export earnings [6] . Agriculture also provided 22.7% of value added in 2014, and an estimated 36.6% of the population is employed in agriculture [7] . Cameroon has great agricultural potential due to its geographic location, which provides an ideal climate for growing cash crops and food crops [8] . Despite this potential, the country continues to spend heavily on food imports. In 2011, the Treasury Department estimated that the government spent nearly FCFA 500 billion (about US$1 billion) on importing foodstuffs such as flour, rice, millet, sorghum, and fish. Therefore, the government needs to address the problems of the agricultural growth sector as it contributes enormously to poverty reduction [9] . The lack of farmer contributions to agricultural production in Cameroon, as in most developing countries, has prompted farmers to seek different options to ensure household food security and maximize farm income. Therefore, in trying to find models for survival and the pursuit of growth, farmers draw their resources from all available places through both formal and informal systems [10] . One way for farmers to increase agricultural production and improve their welfare is to pool and pool their resources to work together as members of Farmer Organizations (FO). FO is defined as “a collective unit of farmers from a village or several neighboring villages, united towards common goals related to economic or social benefits related to agricultural activity” [11] .

Every day, agriculture produces an average of 23.7 million tons of food, provides livelihoods for 2.5 billion people, and is the largest source of income and jobs for poor and rural households. In developing countries, agriculture accounts for 29% of GDP and 65% of jobs [12] ). In addition, there has been an increase in food insecurity, which has been a major problem for farmers’ livelihood in Tubah-sub-division. The prolonged COVID-19 pandemic has delayed the transportation of seeds and fertilizers and thus slowed down the global agricultural process. This affected the productivity of major food crops such as wheat, rice, and vegetables, because agricultural inputs could not be distributed promptly [10] . According to the recent results of the Cadre Harmonisé (2023), 10.6% of people are facing acute food insecurity in Cameroon, which represents 2,940,807 persons [13] . In addition, the mid-year update for the 2023 Global Report against food crises reveals that 22% and 10% of the population are respectively, in phases 2 and 3 of acute food insecurity in 2023. This represents a significant increase in food insecurity, given that in 2022, only 11% of the population was in phase 2, and 10% in 2021 [14] .

Moreover, Low agricultural output has been observed by farmers in Tubah Sub-Division as some of the farmers had to implement some traditional farming methods with the use of local farming inputs such as manure, compost, and traditional seeds such as sorghum, asparagus due to increasing prices and shortages in suppliers of farm inputs which resulted to low agricultural productivity. According to the World Bank (2023), agricultural productivity dropped from 4.02% in 2022 to 2.2% in 2023 [7] . This was due to an increase in farm prices, and shortage of supplies of farm inputs such as fertilizer, pesticides, seeds, weaned animals, and feed coupled with the socio-economic crises in Tubah sub-division during COVID-19 pandemic. Many of these farmers, who belong to organizations like the Bambui farmers cooperatives, whose main farming activities are to cultivate cash crops, do not have access to pesticides and seeds. This is often accompanied by poor farming methods as slash and burn, which greatly reduces the productivity of farm produce in the long-run, hence reducing output. Farm inputs such as chemical fertilizers and insecticides, which could help farmers, improve yields are difficult to acquire by most farmers as their prices continue to increase because of shortages, and flight restrictions impose this increase in prices. In Cameroon, a fall in consumption combined with interruptions to production disrupted global supply chains affecting agricultural industries across Cameroon [15] . The COVID-19 pandemic outbreak in Mezam has caused a lot of businesses to shut down, leading to a monumental disruption of trade and commerce in many agricultural sectors. Farmers and retailers face many short-term challenges relating to workforce, health and safety, cash flow, supply chain, consumer demand, sales, and marketing. Many markets, especially, no longer exist because of these shortages and price increases. As a major effect of COVID-19, the global food prices have linearly increased since February 2020. According to the FAO Food Price Index (FFPI), the international price of food commodities reached the highest level of 97.19 points in September [12] . The value increased by 5% since last year and was the highest between September and February. Similarly, a significant rise was seen in the price index of cereals and vegetable oils for four consecutive months. This was mainly due to the greater shelf life of food commodities [12] .

Purpose of the Study

The Main Objective of the Study is to:

The Main Research Question is to:

The data used in this study were obtained from the survey questionnaire administered to the sample of 202 farmers belonging to FO’s, which comprises common initiative groups and cooperatives. A stratified random sampling method was used to select only participating farmers who belong to a farmer organization. This method was chosen to avoid bias estimates and the problem of simultaneity. The targeted population of the study was divided into strata ( Table 1 ). Firstly, out of the seven Divisions, Mezam Division was chosen because it has the highest number of farmer organisations, which are divided into two: CIGs and cooperatives. These two groups were later categorized into registered farmer organisations and unregistered farmer producer organisations to select individual participating farmers to represent each strata.

Source: ACEFA Mezam Division, 2023.

It is important to note that it is not possible to study the entire population as a result of time constraints and limited resources available for the effective handling of the study. Therefore, only a portion of the population is studied. The opinions and views sampled (a part of population on which the study is focused) from participating farmers in each farmer of the organizations. To ensure the determination of accurate sample size, a sample size of 202 farmers selected from the different farmers organizations.

Model Specification

Objective three assess the effect of sustainability strategies on agricultural productivity in Tubah Sub-Division Northwest Region Cameroon

Olayide et al. (2015) analyzed the policy correlations between agricultural production and sustainable development of agricultural production in Ghana and Nigeria. Emphasizes the influence of political systems and international development programs as correlates of agricultural production and sustainable agricultural production outcomes. This objective follows the argument of reasoning of [23] . Table 2 summarises the indicators of Sustainability Strategies.

Source: Computed by Author (2023).

The dependent variable is agricultural productivity, while the independent variable is sustainable strategies and other covariates.

The model is specified as follows.

$\begin{array}{c}APq={\vartheta }_0+{\alpha }_{1}SS+{\alpha }_{2}AGP+{\alpha }_{3}far{m}_{e}xp++{\alpha }_{4}INC+{\alpha }_{5}PEDU\\ +{\alpha }_{6}SEDU+{\alpha }_{7}TEDU+{\alpha }_{8}Female+{\alpha }_{9}Married+{\alpha }_{10}Member+{\epsilon }_1\end{array}$ (1)

Equation (1) shows the relationship between sustainability Strategies (SS) and agricultural productivity (AP). The parameter ${\alpha }_1$ measures the contribution of sustainable strategies on agriculture productivity at different quantile (q). The magnitude can either be positive or negative. AGP stands for the various age groups and its effect on agriculture productivity at different quantile is captured by the coefficient ${\alpha }_2$. The associated covariate uses as control exogenous variables are summarized on Table 3 .

Source: Computed by Author (2023).

Drivers of Agricultural Productivity

The majority of the farmers had farm size of less than one hectare of land. Obinyan (2000) described the implication of small farm sizes for rural farmers, thus leading to low income and low capital investment [24] . With respect to the income of the farmers, 53% of the farmers earn an annual income of 100,000 FRS - 200,000 FRS and between 201,000 FRS - 400,000 FRS respectively. This translates to average of 12,500 FRS and 25,042 FRS monthly income respectively. Cash crop is the major food production. This accounts for 84.3% of the responses

Table 4 shows a summary of descriptive statistics for socio-economic characteristics of farmers, the observations, mean, standard deviation, maximum, and minimum values. The total distributed questionnaires were 202; among this number, only 198 were returned while 4 questionnaires were missing, giving a response rate 98.02 percent. All the socio-economic characteristics of farmer’s organizations were treated as binary, that is, taking the value of 1, 0 otherwise.

Source: Computed by Author (2023).

Based on the distribution of the respondents with respect to gender, on average, females have a mean of 54.5% and a standard deviation of 49.9%. Meanwhile, male has a mean of 45.5% and a standard deviation of 49.9%. This finding indicates that both female & male farmers were well represented. Balance of opinions is necessary to reduce opinion disparity bias in the study. It further shows that female farmers are more represented than their male counterparts in Tubah sub-division.

According to the age group distribution of respondents, respondents within the age group 30 to less than 40 years have the highest mean of 32.3% and a standard deviation of 46.9%, followed closely by those who are within the age group 20 to less than 30 years (that is, mean of 29.3% with a standard deviation of 45.6%). These age groups are the strongest age groups who can work for long hours on farms; they also have much time to do multi task, and are mostly graduates who can concentrates on farming without distractions from schools. Meanwhile the lowest mean of age group are those from 60 years above and those with less than 20 years with means of 2% and 6.6% respectively and standard deviations of 14.1% and 29.5% respectively, this is because those who are less than 20 years are into school like secondary and tertiary education while those above sixty years are already aging so they cannot work for long hours this called for the reasons why we have small percent of them in farming.

The results in Table 4 . also indicated that out of the total population, the married persons have a mean of 65.2% while the mean for unmarried is 34.8%. This finding indicates some level of social cohesion. More so, it is relevant to know that marital status is a responsibility and stability at individual and community level. Married individuals may be more likely to have grown up in a family with a farming background and continue the tradition. They both have the same standard deviation which is 47.8%

As concerns educational qualification, primary education has the highest mean (33.8%) and a standard deviation of 47.4%, this was followed by tertiary education with a mean of 30.8% and a standard deviation of 46.3%. This implies that much time is tilted to education than farming; that is, people who are into studies give much time preference to education and lesser concentration on farming activities. Those with no education have the lowest mean, which is 13.1% and a standard deviation of 33.9%. This might be because of the inadequate technical skills needed in farming and also because they give more time to concentrate on farming activities than education, which is why the mean is small, and the variation is high.

In addition, Table 4 shows that farmers with farming experience of more than 9 years have the highest mean of 40.4% and a higher standard deviation of 49.2%, while those with 6 - 9 years of experience e and 3 years’ experience have the smallest mean which is 12.6% and 22.2% respectively. The results of the descriptive statistics show evidence of threshold level of experience that farmers need to have for them to be more productive.

Further, the study examines three types of farmers’ organizations, with regards to types of farmers organizations, the CIG’s has the highest mean which constituted 63.1% and a standard deviation of 48.4%, averagely cooperative has 25.8% as mean and 43.8% as standard deviation while associative has the smallest mean of 11.1% and standard deviation of 31.5%. Farmers preferred to work with CIGs than other organizations.

Furthermore, the income group for this research ranges from 100,000 to 1,000,000 FCFA. The above table shows that people with income levels that range from 100,000 to 200,000 has the highest mean (53.5%) and standard deviation of 50% while those with income levels from 601,000 to 1,000,000 have the lowest mean of about 5.1% and the variation is 22%. Farmers with low income may have limited access to modern farming technologies, equipment, and inputs like fertilizers and pesticides. This restricts their ability to optimize crop yields and quality, leading to a wider variation in their production outcomes. Farmers with low incomes often have limited access to education and training programs that can help them improve their farming practices and management skills. As a result, they may struggle to adopt efficient and effective practices, leading to higher variability in their income and productivity.

According to the above Table 5 , based on farm size, it shows that farmer’s with less than 1 hectare of farm size have the highest mean and standard deviation of 49.5% and 50.1% while those with 8 hectares of farm size and above they have the lowest mean and standard deviation of 2% and 14.1%. Comparatively, it shows farmers with 1 hectare have much time and concentrations to their 1 hectare which results in good output with little or no variability in the output as the difference between the variations is 0.6. Meanwhile, there are high variations for those with 8 hectares, showing that there is less concentration in each hectare and that there may be bias. Since they mostly have one hectare of land, the farms mostly practice labour intensive farming with the highest mean (57.1%) and standard deviation (49.6%) than other methods.

With regards to major crops cultivated, food crop is the most cultivated crop with the highest mean (78.3%) and variation of 41.3%. This is because as compared to land fertility, the lands are non-fertile land with its own mean being 55.1%. That is why majority of the farmers mostly obtained quantity of fertilizers from 0 - 50 kgs for their land which constituted the highest mean 33.3% and the standard deviation was 47.3% as seen above. This was followed by 100 kgs of fertilizer with mean (32.3%) and standard deviation (46.9%).

Source: Computed by Author (2023).

Table 6 shows the OLS and Quantile regression result of variables in study; in the 25^th quantile, sustainable strategies has led to a decrease in agricultural productivity. At the 50^th and 75^th quantile agricultural productivity, still declines. This quantile shows the different variations in their mean, despite the mean being the averages of the quantiles. However, these inverse relationships are statistically significant at 1% significant level. Therefore, we reject the null hypothesis and accept the alternative hypothesis, which states that there is a statistical relationship between sustainable strategies and agricultural productivity in Tubah sub-division.

Standard errors in parentheses; ^*** p < 0.01, ^** p < 0.05, ^* p < 0.1.

Table 6 also shows the results for gender; on average females, as compared to males, contribute more to agricultural productivity than males in the 25^th, 50^th and 75th quantile; but this female contribution was highly significant in the 50^th quantile at 1%. This also explains the reason why we have more females in farm organizations than males, they are less occupied than males. Biologically, females are stronger than males, so they can work in an un-conducive environment for a long period than males. Therefore, we reject the null hypothesis and accept the alternative hypothesis, which states that there is a statistical relationship between females and agricultural productivity in Tubah sub-division.

Also, education has a negative association with agricultural productivity. So, Primary, secondary and tertiary education has decreased agricultural productivity in the 25^th quantile and this result is statistically significant at 1% significant level. Meanwhile in the 50^th quantile all these categories of education still reduce agricultural productivity, but only tertiary education was significant at 1%. However, at the 75^th quantile, the Primary and secondary education increases agricultural productivity, though it is not significant, the tertiary education instead decreases agricultural productivity, as seen in the above table.

Moreover, based on farm experience, it shows that those with farm experience from 3 to less than 6 years increase agricultural productivity in the 25^th quantile, though it reduces agricultural productivity in the 50^th quantile, it later increments it at the 75^th quantile. It shows that those with less experience concentrate much time on farming organizations to ensure that their output is better than those with much experience. This relationship was statistically significant at 1% level of significance in the 25^th and 75^th quantile while in the 50^th quantile is at 5%. In addition, those with farm experience from 6 to less than 9 years lead to a decrease in agricultural productivity in all the quantiles, but it is only significant at 10% in the 50^th quantile. On the other hand, farming experience from 9 years above produces a decline in agricultural productivity both in the 50^th and 75^th quantiles, which is significant at 1%, but it was positive in the 25^th quantile.

However, those with a lower income level, from 201,000 to 400,000, have a positive relationship with agricultural productivity, which is significant at 1% in all the quantiles. While those with higher income from 401,000 and above decrease agricultural productivity. This result is evidence that those with low income have a higher marginal propensity to spend on agricultural productivity than those with higher income who have high MPS.

Furthermore, farm organizations such as the associative have a positive and significant effect on agricultural productivity at 5% in quantile 75^th, while cooperatives have a negative and significant effect on agricultural productivity at 1% in quantile 25^th. This implies that the mean of associative organizations in quantile 70 is better than the other quantile.

In conclusion, the study aims to analyze the effect of sustainability strategies on agricultural productivity. Taking a cursory look at all the variables entered in the model has a significant effect. Using the ordinary least square and quantile regression estimation technique, the result revealed that sustainability strategies implemented by farmers reduce agricultural productivity. However, some control variables exert a negative significant effect. All the negative relationships are due to unsustainable farming practices, inadequate capital, low level of education, low farming experience, inadequate income, inadequate farm size, and the type of technology used for farming by these farmers. Based on the findings, this study recommends that the government organize training programs to educate the farmers on sustainable farming practices and implement environmentally friendly policies to boost productivity.

Based on the analysis and findings of this study, the researcher therefore recommends that:

The state should grant agricultural loans to farmers initiate and support mechanized agriculture. This will help improve the productivity and efficiency of farmers in Tubah-sub-division.