Performing Department
(N/A)
Non Technical Summary
The success of small- and medium-sized farms depends on 1.2 million women producers whose numbers grew by 27 percent between 2012 and 2017. This project will fill in a gap in economic research on women primary and non-primary (e.g., spouse) producers on small- and medium-sized farms. We will employ a comprehensive research framework to identify the challenges and growth potential for white and minority women farmers in traditional and emerging agricultural sectors in the Southeast. We will use farm-level data from the Agricultural Census and ARMS to map agricultural industries and niche markets that attract women producers in the region. Subsequent empirical analysis will employ novel econometric techniques to evaluate how access to capital, land, healthcare networks, and government programs affect women farmers' objectives (profitability as well as family obligations), business strategies, and productivity. We will also conduct two regional surveys targeting current and former women producers on small and medium-sized farms to collect more nuanced information to compare the objective Census data with the subjective women farmers' perceptions of their needs and challenges. Our results will provide insights into how access to resources (land, capital, and healthcare coverage), social and environmental factors, and government support programs affect women farmers' operations. Results will also help women improve the efficiency of their operations and offer insights into the impact of COVID-19. This proposal responds to the call to improve the efficiency of operations of small and medium-sized farms and to inform the development of programs for socially disadvantaged farmers.
Animal Health Component
100%
Research Effort Categories
Basic
(N/A)
Applied
100%
Developmental
(N/A)
Goals / Objectives
Long-term goal. The proposed research will help improve white and minority women producers' productivity and access to resources (land, capital and information, as well as healthcare and coverage) and assist policy-makers in designing better support programs that lower the hurdles that women farm operators face.Objective 1. Identify and map agricultural industries and niche markets that attract women farmers (white and minority) on small- and medium-sized farms in the Southeast, separately considering women primary producers (subjectively assigned by NASS rules) and women non-primary producers (self-identified by operators/producers). Alternatively, women producers can be classified as full- or part-time farmers.Objective 2. Evaluate how access to land, capital, information and networks affect women farmer's objectives, business strategies, and productivity using 2017 and 2022 Census of Agriculture Data.Objective 3. Identify to what extent the need for health insurance coverage and access to affordable health care affects women's (primary and non-primary producers) choices to work off- farm part time, and their ability to succeed in their on-farm work.Objective 4. Identify government support programs (federal and state) relevant to women producers in their primary and non-primary producer roles and evaluate the extent to which they help. Identify the consequences of COVID19 using pre- and post-pandemic data and draw lessons for further research.Objective 5. Using insights developed from Objectives 1-4 and Agricultural Census data employ novel empirical models to identify how economic efficiency can be improved for each sub-group of small- and medium-sized operations run by women by utilizing economies of scale and scope.
Project Methods
In collaboration with the contributors, the team will first identify and review existing state and industry-specific studies relevant to women, including studies related to the Covid-19 pandemic. Next, several datasets will be assembled and then analyzed. We will construct a dataset (farm-level) by linking all individual women producers, irrespective of their classification--primary or non-primary producers--in Census years 2017 and 2022 and check if their status has changed from primary to principal small- and medium-sized producers. Separately, the annual ARMS data starting from 2010 will form a supporting smaller-size farm-level women producers panel dataset. The challenge of this data set is that only 3.5% of the sample observations are from women, which has not changed over time, limiting the use of the dataset. Another dataset (county-level) will be assembled from publicly available Census of Agriculture data through Quickstat by NASS, which will become available earlier than the individual 2022 data. Many county-level variables in this dataset that describe the agricultural sector will also be used with the farm-level dataset. Additional county-level data measuring the impact of government payments, weather variability, access to agricultural credit and alternative sources of finance, and agricultural price indexes from a variety of sources will be added. Since women's choice of level of engagement in farming--full-time or in a supporting role--may be determined by environmental as well as family-related circumstances, we will utilize novel data that measure the level of community social capital as well as the inherited beliefs about gender roles (Chetty et al., 2022; McLean et al., 2022). The team at NC State University will also collect data on health insurance and healthcare access from the NAWS as well as data on the state adoption of laws and regulations making farmers eligible for Medicaid health insurance. These may also include CDC data for on-the-job injury rates, disability, and other health-related issues that may affect women farmers' ability to do their jobs, as well as data from the University of North Carolina on rural hospital closures, which affects access to medical care for small- and medium-sized farm operators.Finally, the project will involve collecting our own survey data to meet part of Objectives 2-5 and to juxtapose the objective data analysis with the subjective perceptions of women farm operators' needs and challenges. Since women primary producers are still few in some sectors, a region-wide survey would be more helpful than one focused on a single state.How data will be analyzedDetailed summary statistics data will be generated and used to meet Objective 1. Next, we will employ a range of novel methods for each of the two categories, efficiency analysis and matched comparisons for Objectives 2, 3 and 4. For each group of women farmers and for reference male counterparts, we will estimate the marginal impacts on productivity and profitability ratios of several groups of factors. These factors include economic conditions and opportunities: availability of credit, alternative employment opportunities, input and output price variability, urbanization and distance to major markets, unemployment, and economic growth. Other factors include the use of government support programs, climate and weather variability, and access to health care based on state participation in Medicaid and state-specific rules. Social aspects of women's lives are reflected in social capital and inherited gender attitudes for which we will use appropriate controls (Chetty, 2022; McLean et al., 2022). It is likely that modeling will involve various interactions among these factors. Each of these factors could affect different women producer groups differently.For Objectives 2, 3, and 4, we will also use several impact analysis techniques for the pre-and post-COVID Ag. Census data and for the states with and without the ACA Medicaid expansions that affect off-farm work incentives. These techniques may include Difference-in-Differences and Propensity Score Matching (PSM).Efficiency and productivity analysis (Objective 5) will be used to determine appropriate scale and scope economies scores (from diversification) following Hartarska and Malikov, (2018 & 2019) and Hartarska and Nadolnyak, (2014).7 Novel efficiency analysis will help understand differences between women primary producers, non-primary producers, and men in the same categories.It may be combined with PSM and other matching methods to offer the best possible comparison. Because women farmers are a heterogeneous group, there may be a lack of common support for PSM analysis (observations with propensity score values outside the other group range). This can be addressed by adding variables derived from clustering based on the Principal Components, which improves the covariates' balance in the matched groups (Stuart, 2010; Griffin et al., 2020). Within efficiency analysis, having accounted for firm characteristics and input prices and outputs (within a cost function), the unexplained residuals represent managerial efficiency that may show patterns among different producer categories. One-step stochastic frontier (true fixed effects) models allow to test if the managerial inefficiency is affected by specific factors (as in Hartarska et al., 2014). Unexplained differences in productivity can also be estimated using the Blinder- Oaxaca (1973) and Machado-Mata (2005) approach. We also plan to employ the recently developed four-component stochastic frontier method (Colombi et al., 2014; Kumbhakar et al., 2014) to separate persistent inefficiency from the purely time-varying inefficiency for various women farmer groups, if possible, with relevant ARMS dataset (Objective 5).Data from two surveys targeting (1) current and (2) former women producers (within Objectives 2- 5) will be analyzed with various qualitative dependent-variable methods. The data from the surveys will offer more nuanced data to help juxtapose the objective data analysis with the subjective women farmers' perceptions of their needs and challenges. Since women producers are still few in some sectors, non-white women even fewer, survey data may provide additional details and reveal trends not yet captured by official data. The survey is instrumental because it will compare the women producers' ranking of issues and challenges they face that may be7 Using the North American Industry Classification System (NAICS) or classifying producers by smaller subgroups.different from those identified by the Census data analysis. Current women farm operators will be identified and reached at major farmers conferences (e.g., the Southeast Regional Fruit and Vegetable Conference and the Sustainable Agriculture Conference hosted by the Carolina Farm Stewardship Association), at industry and extension lead growers meetings, and programs focused on women offered by other organizations such as the Farm Bureau's Women's Program. The second survey will target women operators who have exited farming to gain an understanding of which factors (e.g., market, financing, childcare, other family commitments, etc.) and their relative importance contributed to their exit decisions. Work is already underway to identify these individuals. Membership lists for a number of growers' associations and marketing associations are currently being collected. These lists will be recollected again in year 2 of this project to identify those who have discontinued their membership in the intervening years and potentially exited the industry. The survey for former women producers will be sent to these individuals with screening questions to verify that the respondent is a woman who has recently discontinued an ownership role in a farm operation.