For Everything There is a Season: Seasonal Influences on African Elections

• In Event: Poster Session: Comparative Politics of Developing Countries I

Abstract

How does the timing of African elections influence their results? In this paper, I explore how agricultural seasonality influences election results across sub-Saharan Africa and consider how seasonality influences the political attitudes of African voters. Many African voters continue to live in rural communities, working in the agricultural sector. At the same time, only four percent of the continent’s cropland is irrigated. This reliance on rain-fed agriculture ties many African households to distinct rainy and dry seasons, patterning when households engage in different types of agricultural works. Households sow their seeds, weed, and harvest crops at similar times every year. These patterns create predictable demands on household labor, and harvests contribute a predictable influx of earnings.

Research in related social sciences highlights seasonality’s pernicious effects. Every year millions of households enter and exit poverty depending on the time of year. Disease environments vary by season. Diets change by season, both in terms of the quantity of food and the number of nutrients consumed. Local labor markets fluctuate by season. So do relative prices, as the price of food lowers following an increase in supply at harvest, and increases throughout the rest of the year.

Despite significant research in other disciplines, agricultural seasonality receives surprisingly little attention in political science. This paper is one of the first to systematically consider how agricultural seasonality influences political attitudes and outcomes across sub-Saharan Africa. Using Google Earth Engine and 30 years of satellite data from the Landsat program, I construct a measure of agricultural seasonality built on accepted indices from environmental studies. Using the exogenous timing of elections, I combine this measure of seasonality with election returns in a quasi-experiment.

Preliminary results indicate that incumbent parties and politicians benefit at the ballot box when elections are held immediately following harvests. As the time from harvest increases, however, this effect reverses as voters begin to punish incumbents. To better understand potential mechanisms, I pair my measure of seasonality with responses to recent rounds of the Afrobarometer from 28 countries, finding that respondents’ evaluation of the incumbent government’s performance and policy preferences vary systematically by season.

This paper is one of the first to consider the potential political effects of seasonality. It constructs a novel measure of agricultural seasonality relying on over 30 years of satellite data, and pairs this measure with exogenously timed election results and Afrobarometer responses. This paper illustrates how agricultural seasonality does, in fact, influence political attitudes and outcomes across sub-Saharan Africa. Future research should consider what other outcomes may be influenced by seasonality, while future researchers should be mindful of potential seasonal effects when designing their studies.

Author

• Brian Engelsma, University of California, San Diego