Reform Bond Subsidies to Improve States' Behavior

In the last year, I've written extensively about the fiscal holes that states have dug for themselves. State finances have been hit hard by the recession, but many states were fiscally unbalanced even before the economy turned south -- and others have made unforced errors that increased the recession's impact on their budgets. As my colleague Steve Malanga has written, in addition to yawning deficits, this irresponsibility has led to a looming crisis in municipal debt.

Why have some states made such bad choices? How did California and Illinois choose a budgetary path that makes them less creditworthy than Latvia? Part of the problem is sclerotic institutions (especially in California), and partly it's that public officials have too little incentive to care about public finances. But one part of the puzzle is that the federal government actually encourages states to borrow irresponsibly.

The problem is not just that the federal government subsidizes state and local borrowing. Whether local borrowing should be subsidized is a matter for debate -- as is the desirable size of the subsidy, which Congress has arbitrarily set as a percentage of borrowing costs equal to the top personal income tax rate.

But even if bond subsidies are a good idea, there is a key flaw in their current structure: the subsidies grow in value when an issuer's credit profile deteriorates. Because borrowing subsidies are a percentage of interest payments rather than principal, they rise as states are forced by skeptical markets to pay higher yields on new bonds. Even if we keep bond subsidies, Congress should reform the way it subsidizes municipal borrowing to stop rewarding the states with the worst credit.

There are two key ways in which the federal government subsidizes borrowing by states and localities. It has long allowed municipal bondholders to deduct interest income, reducing market interest rates on muni bonds by approximately the top federal income tax rate of 35%.

And in April 2009, it introduced the Build America Bonds program, which provides a 35% interest subsidy on taxable bonds issued by states, allowing them to attract investors who cannot benefit from the tax advantages of traditional munis. Over $90 billion in BABs were issued in the program's first year. These programs are both designed to cover 35% of an issuer's borrowing costs -- and they both provide the greatest subsidies to the states with the worst credit.

To understand the problem, consider the cases of Illinois and Massachusetts. Massachusetts is a relatively creditworthy state, and in May it issued $450 million in long-term Build America Bonds at a 52 basis point spread (0.52%) over 30-year Treasury Bonds. Illinois, meanwhile, has the worst public finances of any state. Its April BAB issuances went at spreads of 205 and 215 basis points over treasuries -- and when it issued again in June, it had to pay a spread of 297 basis points.

These facts reflect the cost of Illinois's messy public finances -- to name a few problems: a decade of unbalanced budgets; the largest FY 2011 budget gap of any state in the country, on a percentage basis; a nearly $5 billion backlog of payments to service providers, chiefly doctors and hospitals who provide Medicaid services; and its arguably worst-in-the-nation funding of its pension system.

To put up with those problems, and the risk of default they create, investors demand roughly an extra 2 percentage points per year in interest to lend to Illinois instead of Massachusetts. But Illinois bears only 1.3 percentage points of that cost -- the rest is passed onto federal taxpayers in the form of Illinois's greater BAB subsidy. Because the treasury pays 35% of interest costs on BABs, regardless of creditworthiness, the program rewards states like Illinois that willfully wreck their public finances.

This doesn't mean that the BAB program needs to be abolished. Instead, the annual subsidy should be a fixed percentage of the principal amount of bonds issued. This could be done in a fiscally neutral way -- for example, the subsidy could be set to match 35% of the average interest rate on BABs issued in a given year. Or, for purposes of simplicity, it could be set at a percentage of the average yield on 30-year treasury bonds.

Let's say for example the fixed subsidy for 2010 issuances had been set at 2% of principal per year. Under this regime, Illinois's effective annual cost of borrowing on its June bond issuance would have been 5.1%, not 4.6%. Meanwhile, Massachusetts's would have gone to 2.7% from 3.2%. This reform would reward states that manage their finances well -- and hopefully encourage other states to improve their fiscal management practices.

Such a reform should not be limited to the BAB market, but should also be applied to traditional munis. This would require a modest complication of the individual income tax code -- instead of allowing taxpayers to deduct muni bond interest, they would deduct a portion of the principal of muni bonds they hold, set to create a tax subsidy equal to the BAB subsidy for top-bracket filers.

In some cases, this would mean a taxpayer would get a deduction in excess of interest received; in others, the deduction would be less than interest received. Such a reform would maintain rough parity in borrowing costs for traditional munis and BABs, and should only be applied to new issuances.

The federal government should consider other reforms to municipal bond subsidies to encourage more sound fiscal behavior at the state and local levels. For example, limits could be established on tax-advantaged borrowing, such as a cap on bond spreads. (If a state can only convince investors to lend it money at an extremely high spread like 300 basis points over treasuries, perhaps the federal government shouldn't be encouraging it to borrow at all.)

Another option is to limit the age of infrastructure subject to bonding, so states don't go out and mortgage old infrastructure to pay for current operations. Malanga points out another piece of low-hanging fruit: Congress should ban the issuance of subsidized bonds to back for-profit ventures, as El Monte, California did to benefit a number of local car dealerships; and as New Jersey is considering for the unfinished mega-mall in the Meadowlands formerly known as Xanadu.

The federal government didn't create the crisis in state and local finance, but it has enabled it. It can stop the enabling by removing incentives for states to ignore their long-term budget problems, and by putting some common-sense limits on the use of tax advantaged borrowing. These may be the first steps on the road to defusing the muni debt bomb.