Abstract
A poisonous and environmentally hazardous byproduct of aviation gasoline combustion is lead emissions. A proposed increase in the aviation gasoline tax from 19.4 to 70.1 cents per gallon has generated a heated debate between pilots and the Federal Aviation Administration. Given that general aviation produces approximately 50% (over 1200 tons) of all lead emissions in the US, understanding the sensitivity that aviation gasoline demand has to price changes is essential to better understanding the policy implications and the environmental impact from an increased tax. Few studies have examined aviation fuel elasticities and no known study has estimated aviation gasoline elasticities. This paper fills that gap and estimates the price elasticity of demand to range from −0.043 to −0.185 in the short-run and from −0.132 to −0.303 in the long-run. Consequently, the estimated impact of the proposed tax increase is a reduction of between 10.71 to 21.30 tons of lead in the short-run and between 15.12 to 34.78 tons of lead in the long-run. Finally, tax revenue could be increased by more than 350% to over $196 million (2009 US$) with the proposed aviation gasoline tax increase.
Notes
1. Gately (1988) finds jet fuel price elasticity of demand to be between −0.10 and −0.15, and Mazraati (2010) finds it to be −0.007, but statistically insignificant.
2. Baltagi and Griffin (1983) utilize consumption per automobile in their methodology of gasoline consumption elasticity analysis. This conversion is utilized given that the per capita consumption may be affected by gasoline prices, and Sterner (1990) determines that this methodology is the most satisfactory in estimating long-run elasticities. Additionally, the number of aircraft that consume aviation gasoline remains fairly steady over the time period studied.
3. Hughes et al. (2008) estimates the price elasticity of demand for automobile fuel consumption between −0.03 to −0.34.
4. To simplify the presentation of the monthly data results in , the coefficients on the monthly dummy variables have been excluded.