Farm size and productivity: smallholder dairy production in Eswatini

ABSTRACT

In response to the 2015 paper by Henderson published In Journal of Agricultural Economics, this case study of dairy farmers in Eswatini, this case study of dairy farmers in Eswatini tests the explanatory power of two hypotheses to explain the inverse relationship between farm size and productivity. To this end, we fit a stochastic frontier production function with inefficiency effects. We find that dairy farmers who use hired labour are significantly less efficient than those who use own and family labour. This supports the labour market imperfections hypothesis. To test the technical efficiency hypothesis, we segment our sample into small, medium and large farmers based on the number of cows in milk. We find that small farmers are the most efficient (78.5%), followed by medium (75.9%) and large (75.1%) farmers, but the differences are not statistically significant. This supports Henderson's finding that differences in efficiency affect productivity but not enough to disqualify labour market imperfections as the principal explanation for the inverse relationship.

KEYWORDS:

• Farm size efficiency
• smallholder dairy
• Eswatini
• stochastic frontier production function

JEL CODE:

• Q12
• Q16

Disclosure statement

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

Notes

1 Observations per sample year: 1998: 4209, 2001: 4191 and 2005: 6879.

2 A Breusch-Pagan test (Breusch and Pagan 1980) test was performed on the model and there is not enough evidence to reject the null hypothesis that the error term is homoskedastic.

3 The median and mean number of cows per group is as follows; small, 2.0 and 1.7; medium, 4.0 and 4.2; large 9.5 and 16.6.