When Taxpayers Ignore Less Visible Taxes

Reminding shoppers of the tax at the time of purchase made for more cautious consumers, suggesting that most of them do not normally take into account the sales tax on such products.

When analyzing tax policies, economists traditionally have assumed that individuals take maximum advantage of (that is, optimize fully with respect to) the incentives created by those policies. In Salience and Taxation: Theory and Evidence (NBER Working Paper No. 13330), Raj Chetty, Adam Looney, and Kory Kroft test that assumption by studying whether the salience (or visibility) of tax rates affects consumers' purchase decisions. The researchers find that salience is quite important in their data, and that salience matters because shoppers are inattentive to taxes. They develop a new theoretical model to analyze the economic effects of taxation with inattentive individuals.

To begin their investigation, Chetty, Looney, and Kroft partnered with a supermarket chain to conduct a three-week experiment in one of its stores. For taxable items, like cosmetics and other non-food products, stores customarily do not include the sales tax in the price tags on the shelves. Instead, the tax appears only on the sales slip when the purchases are rung up at the cash register, making them less salient to the consumer. In the targeted store, the researchers adjusted the price tags to display prices including the 7.375 percent sales tax. The result was a decline in sales of those items by 6 to 8 percent. Reminding shoppers of the tax at the time of purchase made for more cautious consumers, suggesting that most of them do not normally take into account the sales tax on such products.

To complement this experimental evidence, Chetty, Looney, and Kroft ran a second test using observational data over a longer time period and comparing the effect of price changes with tax changes. Here they focused on alcohol consumption, because alcohol is subject to both the (salient) excise tax, which is included in the shelf price, and the less-salient sales tax that appears only at the cash register. Looking at state-level changes in these two tax rates between 1970 and 2003, and at data on annual alcohol consumption by state, the researchers found that the drop in consumption attributable to increases in the excise taxes was measurably larger than the reduction caused by the increases in sales taxes. Thus, sales and excise taxes appear to induce different consumer behavior in both the short and the long run.

The researchers identify two possible explanations for why consumers under react to taxes that are not included in posted prices. One is that shoppers are ignorant of the sales tax rate or of which items are taxable. Another is that salience matters: that is, shoppers know what is taxed, yet still focus on the posted price. To distinguish these explanations, Chetty, Looney, and Kroft surveyed supermarket customers and found that their awareness of the tax rate and of what items were taxable was very high. The researchers concluded that when shoppers make their purchases, they simply do not bother to compute tax-inclusive prices.

To explain this behavior and understand its implications for tax policy, Chetty, Looney, and Kroft develop a theoretical model of inattentive consumers and tax policy. Their first observation is that from an individual's perspective, the cost of not paying attention to taxes is actually quite small. For example, they calculate that the value of learning about a 10 percent tax on an item that costs $1,000 (and thus choosing to spend less on that item) is only $5. Hence, it is not surprising that people who have limited time or attention ignore taxes. Surprisingly, though, the authors show that these same taxes can be quite important from a social perspective: a 10 percent tax raises a substantial amount of revenue for the government, and can create substantial social efficiency costs by distorting economic decisions.

In the authors' model, the tax policies differ substantially from the predictions of the traditional theory that assumes that everyone pays attention to all taxes. A key prediction of the traditional theory is that a tax creates an efficiency cost - that is, a loss of aggregate economic welfare - only to the extent that it reduces demand for the taxed good. In the authors' model of inattentive consumers, a tax can have a substantial efficiency cost even when the demand for the taxed good does not change. This is because a tax that is completely ignored by consumers distorts consumption of other goods. For example, an individual who does not account for taxes on cars would over-spend on the car and end up with less money left than he would like for food or healthcare, reducing economic welfare.

Another implication of the model is that the incidence or pass-through of the tax -- that is, who ultimately bears the burden of paying the tax once price changes are taken into account -- depends on whether the tax is levied on consumers or firms. This result challenges the conventional wisdom that it does not matter if the government taxes consumers or firms, which again is based on the presumption that all individuals pay attention to taxes. The authors give the example of a cell-phone plan whose "sticker price" is $39.99 but whose actual price, including taxes and fees levied on the consumer, may be $47.00. If the same taxes were levied on the firm, it could only pass them through by raising the sticker price, thereby reducing demand. Therefore, firms would be more likely to bear the burden of the tax if it were levied on them rather than on the consumers.

The finding that individuals optimize inaccurately -- even with respect to relatively simple sales taxes -- suggests that similar issues may arise in the analysis of a broad set of government policies. The approach proposed by Chetty, Looney, and Kroft could shed light on a wide range of issues that have received attention in recent policy debates, such as consumption taxation (where taxes may be included in posted prices), social security reform (where the link between taxes paid and benefits received is currently not salient), and the value of simplifying the tax code.

-- Matt Nesvisky