How good are livestock statistics in Africa? Can nudging and direct counting improve the quality of livestock asset data?
Journal of Development Economics · 2025 · Kibrom A. Abay、Hailemariam Ayalew、Zelalem Terfa、Joseph Karugia、Clemens Breisinger
中文摘要
大多数中低收入国家的牲畜统计数据依赖于自我报告的、基于调查的测量方法。然而，受访者在准确报告牲畜拥有量方面可能面临各种挑战。本研究引入了一组新颖的调查与测量实验，以改善非洲的牲畜统计数据。我们对传统的牲畜数据收集方法引入了两项创新。首先，我们通过向随机子集的调查受访者引入明确的助推，来解决牲畜资产中潜在少报的一些来源。其次，我们安排调查员和当地牲畜专家对牲畜资产进行直接清点。我们证明，自我报告的牲畜拥有量数据存在统计显著且系统性的少报。虽然我们的助推仅影响牲畜存栏量较大家庭的报告行为，但直接清点使牲畜总拥有量增加了39%，报告的牛的数量增加了43%。这些影响在牲畜资产拥有的广延边际和集约边际上均很明显，在考虑拥有的牲畜资产数量和价值时也是如此。自我报告的牲畜数据中的这种测量误差可能导致低估牲畜部门对国民经济的贡献。此外，直接清点对调查中未明确清点的牲畜种类产生了重要的溢出效应。我们最后表明，自我报告的牲畜数据中的少报是系统性的，因而对统计推断具有重要影响。我们的发现强调，能够解决自我报告牲畜数据中特定偏误来源的调查设计，可以切实改善非洲的牲畜资产测量。• 本研究引入了一组调查与测量实验。• 我们向随机子集的调查受访者引入了明确的助推。• 我们安排调查员对牲畜资产进行直接清点。• 自我报告的牲畜拥有量数据存在系统性少报。• 直接清点增加了牲畜总拥有量。
Abstract
Livestock statistics in most low- and middle-income countries rely on self-reported, survey-based measures. However, respondents may have various challenges to accurately report livestock ownership. This study introduces a novel set of survey and measurement experiments to improve livestock statistics in Africa. We introduce two innovations to conventional livestock data collection methods. First, we address some of the sources of potential underreporting in livestock assets by introducing an explicit nudge to a random subset of survey respondents. Second, we arrange for direct counting of livestock assets by enumerators and local livestock experts. We demonstrate that self-reported data on livestock ownership suffer from significant and systematic underreporting. While our nudge affects only the reporting behaviour of households with larger stocks of livestock, direct counting increases total livestock ownership by 39 percent and the reported number of cattle by 43 percent. These impacts are evident at both the extensive and intensive margins of livestock asset ownership, as well as considering the number and value of livestock assets owned. Such mismeasurement in self-reported livestock data can lead to underestimation of the contribution of the livestock sector to national economies. Furthermore, direct counting generates important spillover effects to livestock species not explicitly counted in the survey. We finally show that underreporting in self-reported livestock data is systematic and hence consequential for statistical inferences. Our findings underscore that survey designs that can address specific sources of bias in self-reported livestock data can meaningfully improve livestock asset measurement in Africa. • This study introduces a set of survey and measurement experiments. • We introduce an explicit nudge to a random subset of survey respondents. • We arrange for direct counting of livestock assets by enumerators. • Self-reported data on livestock ownership suffer from systematic underreporting. • Direct counting increases total livestock ownership.