Municipal bonds are exempt from the federal income tax, under Internal Revenue Code Section 103. It is the premier tax break available to high income taxpayers, although economists generally view this as to a great extent a subsidy for state and local government, rather than a subsidy for the rich. These bonds are purchased overwhelmingly by top marginal tax bracket taxpayers, and the interest rate paid reflects a discount from comparable corporate bonds that is close to the full value of the tax break received (the discount is more clear for short term bonds than for long term ones).
Never the less, there are problems with the way Section 103 subsidies state and local government. For example, structuring Section 103 as an exclusion from income, rather than as a deduction against income, warps the marginal tax rate structure of the income tax. It makes high income taxpayers with lots of municipal bonds look like lower income taxpayers without municipal bonds. It also isn't very discriminating in what it subsidizes. Stadiums used for the economic gain of privately owned sports teams get the much same benefit as bonds issues to build new schools or a water treatment plant.
Kevin Yamamoto has an interesting new paper at SSRN that explores the current economics of Section 103, reminds us that Section 103 in its current form is not constitutionally required (despite the fact that many people have this misapprehension because the matters was not fully settled law until 1988 when the U.S. Supreme Court resolved the issue in the case of South Carolina v. Baker), and suggests one possible reform. His reform proposal is nothing to write home about it, although it has some good features, but his criticisms of Section 103 as it exists today and his analysis of the state of the law and economics of the municipal bond exclusion from federal income tax is insightful.
Yamamoto is clearly a hardworking law prof on the make in the field, writing about a variety of interesting policy subjects.
Hat Tip: Tax Profs Blog.