The Effects of Educational Externalities on Maize Production in Rural Malawi

Abstract

The paper looks at the existence, nature and form of intra- and inter-household externalities of education on productivity, efficiency and uncertainty of maize production in rural Malawi. Data from the Third Integrated Household Survey are used. I find statistically and economically significant positive intra- and inter-household externalities from education on all three elements, and that intra-household externality effects are larger than inter-household externality ones. Community-level schooling is found to substitute for household-level schooling in the sense that farmers who reside in households where members are not educated nevertheless have relatively higher production and lower production uncertainty, on account of living in communities where some inhabitants are educated. The paper also finds that the intra- and inter-household externality effects are more pronounced for the least efficient farmers, that they are monotonic and that they are largest when average household schooling is relatively low.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

¹ In this study, technical inefficiency is defined as the deviation between observed and maximum feasible output for a given production technology, and observed input use, i.e. a production frontier. Farmers can either operate on the frontier if they are technically efficient or beneath the frontier if they are not technically efficient. Productivity, on the other hand, refers to the relationship between observed output and observed input use for a given level of production technology. Here farms are assumed to be operating at full technical efficient levels and thus do not purposely waste resources. Any resource wastage is attributable to ignorance.

² Alternatively, community-level fixed effects can be used here; however, since there are 624 communities after data cleaning, this means estimating too many fixed effects, and a loss of degrees of freedom.

³ The average output elasticities as calculated using the following formula:

where is the sample size.

⁴ A complete set of results is available from the author upon request.