Fare evasion in public transport: A time series approach

Abstract An econometric model is presented that identifies the main variables explaining evasion of fare payment on a public transport system. The model uses a cointegration approach. The model parameters are estimated using data from the Santiago (Chile) bus system, where evasion has been measured at approximately 28%. The main results of the model are that (i) a 10% increase in the fare raises evasion by 2 percentage points and (ii) a 10% increase in inspections lowers evasion by 0.8 percentage points. An increase in unemployment, the third explanatory variable in the model, tends to induce a decrease in evasion, and vice versa. This counterintuitive finding may be explained by the fact that those most vulnerable to job loss, and more likely to evade than the average user due to economic necessity, tend to reduce their use of the bus system when unemployment rises and increase it when unemployment falls. Our results suggest a revision of the evasion control policy in Santiago to improve its effectiveness, and to link inspection efforts to fare increases or to decreases in unemployment.