On the Complexity of Eliminating Fuel Subsidy in Indonesia: A Behavioral Approach

People’s attachment to a subsidy creates difficulties for the government to phase out, and eventually eliminate, the subsidy. Elimination of fuel subsidy scheme in Indonesia is a perfect example of such occurrence. The subsidy has been implementing to commodity as opposed to households, thus individuals may not necessarily realized that they have been enjoying the subsidy when they buy fuel. In this case people may feel as if they are endowed by the values from the provision of the policy. The elimination of the subsidy consequently may be perceived as a loss - as opposed to a foregone gain. This study aims to obtain the most acceptable exit strategy to eliminate the subsidy from the perspective of households by conducting a laboratory-based survey. The alternative exit strategies include methods of elimination of the subsidy and of reallocation of resources saved from eliminating the subsidy. The policy options have been derived using insight from behavioral economics ranging from endowment effect, status quo bias, to present biasness. The survey includes 335 subjects, who come from four different backgrounds: 1) households with no motor vehicle; 2) households with only motorcycle(s); (3) households with one car and; 4) households with one luxurious car or more than one car. Each subject faces 55 paired-wise policy alternatives and the method proposed by Dunn-Rankin (1983) has been used to derive the ordering of preferences. The result shows that gradual elimination of fuel subsidy and reallocation to earmarked programs were the most acceptable policy elements of the exit strategy. The survey, indirectly, showed that subjects’ valuation of losses is greater for direct elimination strategies than that of the equivalent gradual elimination strategies. The results also show that respondents chose “to pay” later at a smaller amount than “to pay” immediately of the equivalent total value. The reallocation of resources saved to earmarked programs is more acceptable than the reallocation to non-earmarked programs. In particular, respondents opted for a more immediate compensation from the elimination or reduction of the subsidy.