Recent years witnessed substantial growth in sales of US organic food [1] [2] [3]. The annual growth rate of organic food sales between 1990 and 2017 reached a double digit ranged from 12% to 21%, far outpaced the growth of products conventionally produced [2]. In 2106, the total sales of organic products had grown to $47 billion [4].

The fast-expanding US organic market was partially attributed to the strong demand due to an increasing number of consumers’ preference for buying and consuming organic foods. The niche market had embraced as high as 69% of US consumers [5] and they were willing to pay the premium prices for organic products [6] [7] [8] [9] [10]. The driving forces behind include growing concerns about health, valuing food safety, environmental shepherd, and animal welfare [11] [12]. The conversion to organic foods was particularly significant in consumers’ favor for organic produce [7] [10] [13].

While the solid demand for organic products unfolded in both domestic and international food markets, the supply side revealed a relatively stagnant growth within the United States. Overall, organic production has not kept pace with the growth in demand, organic farms are struggling, and organic handlers are challenged for procuring enough [7] [14]. At the very upstream of supply chain, organic farming remains a tiny share, accounting for about 0.55% of the total US farm land [15].

The shortage of supply along the organic chain has been much in evidence for years in the United States. When it comes to the major barriers, studies point to low yields, higher costs, unstable product quality, the cost of certification, price premiums, imports, limited organic inputs, and the lopsided competitions [6] [16] [17] [18]. Some others render explanations with focuses on psychological and sociological costs [19], on uncertain legislative environment [1], on security of marketing channels [20] [21], on input constraints [22], and the perceptions of farmers [23]. While the varieties of investigations did cover a broad area, the results are mixed. The lack of consensus regarding the influential factors and their impact on organic conversion calls for more quantitative analyses based on good survey data, and some studies based on econometric models step forward to quantitatively address the choice of farmers on the conversion to organic production and reached conclusions that related organic farming to the age and gender of operators, farming acreages, levels of education, and urban-raised farmers, which did not fully resonate with some early analyses, but provided a more informative vision [21] [24] [25] [26]. Despite the progress, a closer look at the limited studies finds a problem of small sample size, which is linked to problematic estimates of impacts in spite of their significance in statistics. This is a well-recognized issue to many, but has a tendency to repeat itself due to the predicament in organizing interviews and collecting data. Consequently, conflicts remain even in this category of studies. The situation merits more studies with the well-designed survey instruments and relatively large samples to ensure more accurate and consistent estimates of the impacts of influential factors in the small-farmers’ decision-making on the conversion to organic production.

This study has its focus on the choice of small farmers in the conversion to organic production and is aimed to identify the factors of the influence and validate their impacts on farmers’ decision on the conversion to organic production. Based on a large survey data in the Southern region of the United States, the authors examined farmer’s attitude, perception, farms characteristics, features of the local community, social demographic factors, and so forth. The results supplemented some conclusions of early studies and offered a deeper understanding of choices of farmers in the conversion to organic farming in the United States. A broad survey and econometric modeling featured this study make the analysis ready to the empirical application and bring about the important implications for understanding the complex organic conversion process and for making workable and effective polices to support the small farms in the United States.

The data used in this study came from a broad interview with small farmers in Southern region of USA in 2014. The information collected covered organic practices adopted, societal and demographics of operators; characteristics of farms; farmers’ beliefs, attitudes, and perceptions related to organic farming practices, and so forth. A large sample of farmers from a population of total 4818 farms in Georgia were randomly selected and interviewed. The instruments used in the interview were designed by faculty members of Fort Valley State University and the field work of the survey was administered by the Burruss Institute of Kennesaw State University. Farmers who respond to the screening question, “Do you produce fruits and vegetables?” with a “yes” were retained in the sample. The qualified interviewees were contacted multiple times to obtain an effective sample size. About 2404 farmers were contacted, with 456 farmers going through the survey, which gave a return rate of 18.9%. The instruments encompass a broad spectrum of questions pertinent to production practices, social demographics, individual attitude, beliefs, perceptions, as well as the characteristics of farms. The organic production in the survey may take the form of the USDA certified, certification exempt, or transitioning farms. The interview were conducted by trained personnel following the well-established procedures, which insures the veracity of data collected. However, it is also in evidence that some self-selection biases occurred due to the fact that the higher level of education were associated with organic producers and they were inclined to finish the survey retained in the sample, which make the organic operations in the sample high than the overall percentage in Georgia farms in 2012 Census of Agriculture. The defect may constrain the effort to reach a general extrapolation beyond the survey data. In Table 1, the variables covered in the survey and the corresponding preliminary statistics were reported to provide a profile of small farmers in the Southern region of states.

We approached farmers’ choice of organic farming and potential factors of influence with the help of the logit regression model. After comparative study on the logit, the probit, and the linear probability models, being alike in ways in analyzing categorical data [27] [28] [29], we first exclude the linear probability model for its bias and inefficiency [28] [30]. The logit and probit models are

*The percentage sum of each variable may not equal 100 due to the responses of unknown or inapplicable information.

arguably equivalent, only many investigators prefer the former for easy interpretation of parameters. For the sake of comparability, we used the logit model hereof. Since rich documents related to the logit model are readily available in the literature [31] [32], the authors just present a model brief in the context of this investigation, rather than a thoroughgoing model discussion in the coming section.

Following [31], we assumed a random variable Y_i, taking one of three discrete values indexed as 1, 2, 3, representing three operational choices: pure conventional, mixed, and pure organic operation respectively. Let

P i j = Pr ( Y i j ) (1)

Denotes the probability that the i^th response choice falls in the j^th category. So, P_i1 represents the probability that the farmer i choose to do conventional farming. It was assumed that log-odds of choice j of respondent i with reference to the choice 1 follows a linear relation as described in Equation (2).

η i j = log p i j p i 1 = β 0 + X ′ β j for j = 2, 3 (2)

where β₀ is a constant, and β_j is a vector of regression coefficient for j = 2 and 3.

The model above is analogous to a simple logistic regression model, except that the probability distribution of the response is multinomial instead of binomial and two equations are estimated instead of one. The equations contrasted each of categories 2 and 3 with the choice 1, whereas the simple logistic regression is just a contrast between success and failure. In the support of SAS software, two models were fitted with the likelihood maximization method. The coefficient vector β_j is associated with a set of explanatory variables and measures the change in log odds for a one-unit increase in X_i. The odds ratio derived is the change in ratio of the odds for a one-unit change in X_i [33].

In order to have an easier and intuitive interpretation of the impact from each factor, the probability of each choice was estimated. Then, the marginal impact of each factor on probability of a choice were able to be derived from the three equations below:

Pr ( Y = 1 ) = 1 e X β ( 1 ) + e X β ( 2 ) + e X β ( 3 ) (3)

Pr ( Y = 2 ) = e X β ( 2 ) e X β ( 1 ) + e X β ( 2 ) + e X β ( 3 ) (4)

Pr ( Y = 3 ) = e X β ( 3 ) e X β ( 1 ) + e X β ( 2 ) + e X β ( 3 ) (5)

As usual, the most challenging part of modelling is associated with model selection among many alternatives. In this study, we adopted the approach of Purposeful Selection of Variables [34], which usually retains important confounding variables and result potentially in a slightly richer model. We began our model fitting by a univariate analysis of all variable relevant. Any variable with a significant univariate test at 0.25 level was selected as a candidate for the multivariate analysis. In an iterative process, covariates are removed from the model if they are non-significant and not a confounder. Significance is evaluated at the 0.1 level and confounding as a change in any remaining parameter estimate greater than 15% as compared to the full model. A change in a parameter estimate above the 15% indicates that the excluded variable was important in the sense of providing a needed adjustment for one or more of the variables remaining in the model. At the end of this iterative process of deleting, refitting, and verifying, the model contains significant covariates and confounders. Then, we took into account of any variable not selected for the original multivariate model and added them back one at a time, with all significant covariates and confounders retained earlier. In such a way, other variables which, by themselves, were not significantly related to the outcome but became an important contributor in the presence of other variables will be included in the final model. All variables that are significant at the 0.1 level were put in the model, and the model was iteratively reduced as before but only for the variables that were additionally added. Following the steps above, we obtained a final version of model as organized and presented in Table 2 and Table 3. The former is in terms of log odds and the latter in terms of probability.

*Note: Triple, double, and single asterisks indicate significance at the 1%, 5%, and 10% level, respectively. Mixed = organic and Conventional, Pure = all organic, and Con = Conventional, Likelihood Ratio (Pr < ChiSq) < 0.0001.

Note: dy/dx for factor levels is the discrete change from the base level. _predict 1 is the probability of conventional farming; _predict 2 is the probability of the mix of conventional and organic farming; _predict 3 is the probability of pure organic farming.

The fitted models reported in Table 2 and Table 3 are significant at the 1% level in terms of AIC, BIC, and Score statistics. Ten variables were retained at the 10% significant level, reflecting the responses to the following survey questions: “Do you use farmers market as a marketing outlet?”; “Do you find reliable buyers for your organic products?”; “Do you start organic farming by following some successful peers?”; “Do you believe organic farming is more prone to lower yield than that of conventional farming?”; “Do you believe that more liability incurred in organic farming?”; “Do you have an off-farm job?”; “What is the size category of your farm?”; in addition, questions related to age, years of farming experience, and years of education of the major operator were identified as independent variables of influence in the decision making process of farmers. The following discussion will be on important impact of each identified factor in the finalized model.

Analysis of the large survey in the Southern states came up with a few contributing factors in farmers’ choice of conversion to organic production. The pool of factors comprise barriers and stimuli, each of them have been discussed in detail in the previous section. In view of potential impacts and the efficient way of improve the adoption of organic farming, the four factors of risk aversion, the age of operators, and the size of farms, and marketing channels deserve further elaboration.

Organic farming is still at its early stage, there exist tremendous uncertainty in both production and marketing process. For most risk version farmers, an acceptable way to follow was doing by the top dog. Nevertheless the exemplars available at this stage are quite limited. A way to increase farmers’ expose to and know those successful frontrunners is to organize more workshops and training. Land grant universities and USDA agencies such as FSA and RMA could play a more active role in helping farmers overcome the risk barrier.

The age of operators and its adverse impact certainly deserve multiple looks. From 2012 census data, about 62% of farmers were 55 or older, which was 17 years older than the age of average American workers. Old operators was significant less likely to go organic. While it isn’t realistic to expect a change of age structure shortly, it makes sense to support young farmers in land requisition and in tax relief. Currently, the old operators’ ownership of expensive land and the existing subsidy programs give the elder an edge to stay in farming, which forms a hindrance of the conversion to organic farming. An urgent need is to have some stimulus mechanisms to expedite the influx of the young into farming business. The change of land tenures and government subsidy programs is needed to support the business of the young beginning farmers

The third noticeable factor is the size of farm. The last century witnessed the significant land consolidation in the US agriculture. “Be large, or be out” is almost a common knowledge to all farmers. The high percentage of capital and land was under control of large farms makes the conversion to organic farming a predicament process. Increasing the organic production may hinge on the slowdown of the farmland consolidation, which may need government to steer some subside programs partially away from just a few large staple crops.

In the end, another factor, which should not be ignored, is the demand paradox along the organic food supply chain. While substantial evidence pointed to a great aggregate organic market, organic farmers did that way. Having reliable buyers is still a great concern for many small farmers. It demonstrated that access to farmers markets is vital, but not enough. Going through alternative marketing channels, such as the big box chain and food hubs is an inevitable step to ensure a stable gain in the conversion to organic farming. More studies on the way of channeling organic products to the big markets are merited in order to help producers go organic and make a living.

The authors declare no conflicts of interest regarding the publication of this paper.

Liu, X.L., Pattanaik, N., Nelson, M. and Ibrahim, M. (2019) The Choice to Go Organic: Evidence from Small US Farms. Agricultural Sciences, 10, 1566-1580. https://doi.org/10.4236/as.2019.1012115