What do one hundred million transactions tell us about demand elasticity of gasoline?

The price elasticity of gasoline demand is a key parameter in evaluation various policies. However, most the literature uses aggregate data to identify this elasticity. Temporal and spatial aggregation make such estimates biased. We employ unique dataset all transactions Iran during 4-month period around an unexpected exogenous change that also significant withholding behavior by consumers response anticipated changes. reduce or postpone their purchases when they expect decrease prices. Controlling for date fixed effects would eliminate homogeneous responses. heterogeneous responses lead overestimation After controlling date, individual, location as well behavior, we estimate robust − 0.085. Aggregation same week, month, city yields 0.3, indicating bias earlier studies.