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.
From NBER Digest;
Every year, as April 15 approaches, taxpayers must take the time to calculate - and then pay - their federal and state income taxes. Indeed, economists have estimated that for every dollar paid in taxes, taxpayers incur an additional 10 cents in time costs associated with record keeping and tax filing. Many policymakers and economists have conjectured that time spent paying taxes is important for keeping taxes visible and salient to taxpayers, thereby making it politically harder for the government to raise taxes.
In E-ZTAX: Tax Salience And Tax Rates (NBER Working Paper No. 12924), NBER Research Associate Amy Finkelstein investigates this conjectured link between the visibility of taxes and the level of taxes. She studies the impact of electronic toll collection systems - such as E-ZPass in the Northeast or Fast-Trak in California - on toll rates. Because these electronic systems automatically deduct the toll as the car drives through the toll plaza, and the driver therefore need no longer actively count out and hand over cash for the toll, electronic payment arguably reduce the visibility of tolls.
Finkelstein finds that this reduced visibility of tolls comes at the cost of higher tolls. She estimates that the introduction of electronic toll booths causes drivers to pay higher tolls - some 20 to 40 percent higher - than if electronic collection had never been introduced.
For her study, Finkelstein collected 50 years of toll data on 123 publicly owned roads, bridges, and tunnels in the United States. Starting in 1987, electronic tolling was introduced on these facilities. By 2005, about two-thirds of the facilities used electronic tolling. Once a facility introduces electronic tolling, drivers start to use the technology, and eventually usage levels out at about 60 percent of toll payments.
Finkelstein finds that as drivers switch to paying tolls electronically, toll authorities raise the toll rates. As a result, even though many facilities offer discounts to drivers who pay electronically, the toll that drivers end up paying electronically is still higher than it would have been had the facility not introduced electronic tolling (although it's lower than what their fellow drivers who still pay with cash have to fork over!)
The most plausible explanation for the phenomenon, Finkelstein argues, is that drivers who pay the toll electronically don't notice price hikes as readily as manual-toll users do. So public resistance to toll increases lessens as more and more drivers pay electronically, and thus transportation authorities are able to push through more toll increases.
Automated tolls, after all, are fairly hidden. A driver only has to slow down so that her car's ID tag can be scanned and the toll automatically deducted from her account. When her balance falls below some preset minimum (typically $10), the transportation authority automatically debits her credit card or bank account. Small wonder, then, that survey evidence shows that drivers who pay electronically are much less aware of how much they have paid than drivers who pay using cash. Also supporting the "decreased visibility hypothesis", Finkelstein finds that traffic decreases less in response to toll increases when a larger share of the tolls are paid electronically (rather than in cash).
The study examined other possible explanations (than the decline in toll visibility) for the increase in tolls when the use of electronic tolling rises. For example: drivers may like the convenience of paying electronically so much that they're willing to pay more for it. But that thesis didn't hold up in two telling instances, Finkelstein says. First, on roads where manual tollbooths really slowed drivers down, the change to electronic tolling saved them much more time. Yet, these roads did not see unusually high price increases. Second, drivers saw similarly large savings of time when bridges and tunnels switched from charging tolls in both directions to charging tolls in only one. But again, the toll increases were not out of line with the norm.
Other possible explanations - that toll authorities had to raise rates because of the costs of installing the automated system or that they used higher rates to battle congestion or recoup revenue - didn't hold up either to the evidence, the study found. That leaves the original conclusion as the leading explanation: the more hidden the tax, the less resistance it breeds, and the easier it is for governments to raise taxes.