The political economy of farmers’ suicides in India: indebted cash-crop farmers with marginal landholdings explain state-level variation in suicide rates

Abstract

BACKGROUND: A recent Lancet article reported the first reliable estimates of suicide rates in India. National-level suicide rates are among the highest in the world, but suicide rates vary sharply between states and the causes of these differences are disputed. We test whether differences in the structure of agricultural production explain inter-state variation in suicides rates. This hypothesis is supported by a large number of qualitative studies, which argue that the liberalization of the agricultural sector in the early-1990s led to an agrarian crisis and that consequently farmers with certain socioeconomic characteristics-cash crops cultivators, with marginal landholdings, and debts-are at particular risk of committing suicide. The recent Lancet study, however, contends that there is no evidence to support this hypothesis. METHODS: We report scatter diagrams and linear regression models that combine the new state-level suicide rate estimates and the proportion of marginal farmers, cash crop cultivation, and indebted farmers. RESULTS: When we include all variables in the regression equation there is a significant positive relationship between the percentage of marginal farmers, cash crop production, and indebted farmers, and suicide rates. This model accounts for almost 75% of inter-state variation in suicide rates. If the proportion of marginal farmers, cash crops, or indebted farmers were reduced by 1%, the suicide rate--suicides per 100,000 per year--would fall by 0 · 437, 0 · 518 or 0 · 549 respectively, when all other variables are held constant. CONCLUSIONS: Even if the Indian state is unable to enact land reforms due to the power of local elites, interventions to stabilize the price of cash crops and relieve indebted farmers may be effective at reducing suicide rates.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 99 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.