A basic tenet of public finance holds that people tend to do less of something when it is taxed.  Raise income tax rates and some people will work less. Boost the gas tax and people will drive less. Hike the cigarette tax and people will smoke less.

That inexorable law of demand poses two problems for the taxman. First, taxes distort behavior as people move from taxed activities to those that are taxed less or not at all. Sometimes, as in the case of cigarette taxes, we want to discourage the taxed activity. In other cases, the tax only makes the economy less efficient.  Second, tax avoidance may reduce the revenue gained from a tax increase—or even negate it entirely.  For example, if gasoline sales plummet when gas taxes rise, we get less revenue to build and maintain roads.

But recent research suggests that taxes don’t always have to depress demand. People may not react to tax changes they don’t perceive. If the price change isn’t obvious, homo economicus goes on merrily consuming the same as before.

MIT economist Amy Finkelstein examined the behavior of motorists using toll roads with and without electronic toll payments. Because drivers don’t fork over cash to pay tolls when they use electronic transponders, they are less aware of the cost and their demand is less responsive. Toll revenues, Finkelstein found, are 20 to 40 percent higher with electronic toll payments than under the old cash-only system.

Raj Chetty, Adam Looney, and Kory Kroft compared consumer behavior when sales taxes were included and excluded from marked prices and found an 8 percent drop in demand when people saw the tax-inclusive price on the shelf instead of having the tax added at the cash register. The salience of taxes clearly matters: people don’t react to taxes they don’t see.

Tax complexity might also reduce tax awareness. For example, many taxpayers have no idea whether they will owe the alternative minimum tax (AMT) until their accountant or TurboTax tells them. If they don’t know it’s there, they won’t change behavior and the government doesn’t lose needed revenue. It is the same with the phase-outs of the personal exemption (PEP) and itemized deductions (Pease). They are effectively rate increases that few notice.

Of course, complexity can work the other way too. The energy conservation tax credits in this year’s stimulus bill subsidize the purchase of various “green” products that save energy, ranging from low-e windows to high-efficiency air conditioners. But, as Rosanne Altshuler has pointed out, it’s not easy to figure out whether a specific item qualifies. Sellers will no doubt advertise their qualifying products but more consumers might take the bait if the tax rules were simpler. Similarly, lots of evidence suggests that people don’t take full advantage of tax-deferred savings accounts because of confusing rules and the wide variety of choices available.

Furthermore, taxpayers may be so confused by the rules that they respond in perverse ways.  Some people think that the phaseout of itemized deductions actually reduces the value of additional charitable contributions and mortgage interest deductions.  It doesn’t.  AMT taxpayers may be confused about what is deductible and what isn’t.

I’m not arguing in favor of complexity. Taxpayers don’t trust a tax system they don’t understand. Complexity may reduce compliance, either because people have trouble following the rules or because they think others must benefit from obscure provisions and they should somehow pay less tax too. And we often do want tax provisions to affect people’s behavior.

Nonetheless, complexity may sometimes mask taxes and thus help to raise revenues in a relatively efficient way.