Are the Income Gains from a Development Project Consumed or Saved?

We use a matched difference-in-difference design to estimate the time profile of the impacts on household consumption and income of an aid-financed poor-area development project in rural China. Sampled households in project and matched comparison villages were followed up annually over the project’s disbursement period. We find that beneficiaries saved about half the total income gain attributed to the project — well above their average saving rate. This is consistent with the high year-to-year variability we find in the project’s impact, which would have made it hard for participants to infer the gain in permanent income.