This paper tests the hypothesis that the salience of a tax system affects equilibrium tax rates.To do this, I analyze how toll rates change after toll facilities adopt electronic toll collection. Unlike manual toll collection, in which the driver must hand over cash at the toll collection plaza, electronic toll collection automatically debits the toll amount as the car drives through the toll plaza, thereby plausibly decreasing the salience of the toll. I find robust evidence that toll rates increase following the adoption of electronic toll collection. My estimates suggest that, in steady state, toll rates are 20 to 40 percent higher than they would have been without electronic toll collection. Consistent with the hypothesis that decreased tax salience is responsible for the increase in toll rates, I also find evidence that the short run elasticity of driving with respect to the actual toll declines (in absolute value) following the adoption of electronic toll collection. I consider a variety of alternative explanations for these results and conclude that these are unlikely to be able to explain the findings.Key Words: tax salience; size of government; tolls, electronic toll collection JEL Classification Codes: H11, H71, R48 I am grateful to Daron Acemoglu, David Autor, Raj Chetty, Peter Diamond, Liran Einav, Mike Golosov, Erzo Luttmer, Sean Nicholson, Ben Olken, Jim Poterba, Nancy Rose, Stephen Ryan, Monica Singhal, and seminar participants at Cornell, MIT, Berkeley, and Stanford GSB for helpful comments, to Julia Galef for outstanding research assistance, and to the innumerable employees of toll operating authorities around the country who generously took the time to provide data and to answer my many questions 1