A ‘cultural transformation’ at the US Department of Agriculture?: Examining racial (in)equality through federal farmland protection programs in Georgia

ABSTRACT

Over the past several decades the US Department of Agriculture (USDA) launched a 'Cultural Transformation Initiative' aimed at addressing racial discrimination. Yet recent work from a wide range of sources questions whether these efforts have resulted in meaningful material transformations to USDA governance. This article focuses on one of the USDA's main farmland protection programs and analyzes the extent to which it challenges and/or reproduces racial inequalities in the state of Georgia. We conclude that the program continues to present significant barriers to racial equity, many of which stem from national-level criteria. Moreover, the USDA's internal civil rights audit mechanism also fails to address or acknowledge these problems. We use quantitative and qualitative methods to highlight how racial inequality is reproduced through class biases, and argue that any meaningful transformation in the uneven effects of USDA programs requires attention to the historical geographies of land ownership.

KEYWORDS:

• Dispossession
• USDA
• conservation easement
• racial capitalism
• discrimination
• land

Acknowledgements

The authors thank Arundhati Jagadish for her helpful discussions and comments on earlier drafts. They also thank the anonymous reviewers whose comments greatly improved the final manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 See, for instance, Alonzo (2017).

2 To assess the full value of the easement, an appraiser determines the difference in a property's fair market value before and after the easement restricting the landowner's development rights.

3 The CRIA defines ‘protected groups’ as ‘women, minorities, and persons with disability’.

4 For useful reviews of this literature, see Melamed (2015); Burden-Stelley (2020); Jenkins and Leroy (2021).

5 Goldberg (2002) argues that race is integral to the conceptual, philosophical and material dimensions of modern nation state formation and management.

6 The 2501 Program defines ‘socially-disadvantaged group’ as ‘a group whose members have been subjected to racial or ethnic prejudice because of their identity as members of a group without regard to their individual qualities.’ An additional complexity of this definition is that under some USDA programs gender identity may be considered for inclusion in a ‘socially-disadvantaged group’ and under other programs it is not.

7 Analyses of parcel-level criteria using county-level data would mask within-county variation on the distribution of parcels owned by farmers of color across the different ranking criteria.

8 Pearson's correlation coefficient measures how well two numeric variables are related. Its possible values range from r = −1 to 1. r = 1 indicates a perfect positive correlation, r = −1 indicates a perfect negative correlation, and r close to 0 (generally within 0.3) indicates little to no correlation.

9 The USDA Ag Census does not sample every farmer in the United States like the decennial census of people in the country does. The tallies for the total number of farmers by race are estimates. It is worth noting that the USDA estimates only about half of all farmers of color respond to the Ag Census and that it uses weighted calculations to interpolate the remaining statistics (USDA 2017a). Thus, the total number of farmers of each race could vary substantially from the numbers reported here and in all tables below that report data from the Ag Census (also see Garcia, Lopez-Ariza, and Marinez 2008; Garcia and Marinez 2005).

10 P-value is a measure of the probability of obtaining observed results assuming that distributions are random. In our case, if race had no effect on who was awarded FRPP easements and easements were awarded randomly, our p-value of <0.00001 means that we would expect to see similar results less than once out of every 100,000 iterations. This means that our results are extremely unlikely in such a scenario. Because of that, we reject the notion that race has no effect on who gets FRPP easements. Typically, statisticians accept 5 out of 100 (p < 0.05) as statistically significant.

11 In power analysis, the level of power indicates the probability of avoiding a Type II error, of failing to reject a false null hypothesis.

12 Under the 2014 Farm Bill, the cost of the easement had to be split between the NRCS (50%), the landowner (25–45%) and an eligible third party (5–25%), typically the local or state government. The tendency in practice was for the landowner and third party to make equal contributions at 25%. While the landowner supplies her/his share through a charitable partial donation, the third party covered their portion of the balance through a cash match (Brown et al. 2017). This served as a serious constraint on program participation under the old Farm Bill (Oct 2020 interview with state NRCS representative), which in Georgia had been largely limited to two relatively wealthy counties where green space was valued and citizens supported the use of local sales tax revenue for this purpose.

13 With the analysis focused exclusively on Georgia, this conclusion should not be interpreted as a general trend.

14 The NRCS categorization of ‘historically-underserved’ explicitly excludes consideration of gender or sexuality for protected group status. See USDA (2022b). As with other USDA programs, if an applicant falls in more than one of these protected groups they would not gain more than 30 points in their final ranking score.

15 Seventy-five of these points are not included in Table 3, as they are entity-specific criteria (e.g. pertaining to the qualifications and performance of the land trust or other entity holding the easement).

Additional information

Notes on contributors

Levi Van Sant

Levi Van Sant is a human geographer who studies the environmental justice dimensions of agriculture and conservation, particularly in the US South.

Laura German

Laura German is a cultural and ecological anthropologist who studies land and environmental governance. Her recent work, informed by critical agrarian, legal and development studies, explores the intersections between political economy, law and knowledge/ontology in the international land governance arena.

Daniel J. Read

Daniel J Read is an environmental anthropologist who studies the politics and ecology of conservation in working landscapes in India and the United States.