Originally posted at Public Sector Inc.
By Dan Rothschild.

Windfalls are, by definition, irregular occurrences, which disrupt state and local governments’ regular flow of finance. And because each windfall is, at least to some degree, sui generis, state and local policymakers can have a rough time of figuring out how to account for windfall money and how to spend it.

The tobacco master settlement agreement (MSA) of 1998 is perhaps the ur-example of how windfall funds can be misappropriated and spent on any number of things totally unrelated to the putative purpose of the windfall–and wreck havoc on budgets for generations to come.

The problem is not merely that it is some kind of moral betrayal to spend windfall revenues on things unrelated to the source of the windfall. Rather, it is that spending windfall money on anything but one-time expenses–that is, either capital improvements or remedies for whatever problems led to the windfall–can grow public payrolls and taxpayer obligations in a way that requires either tax hikes or service cuts in the future.

It’s instructive to take a step back and think about how we should conceive of windfalls for public budgeting purposes. While there are numerous sources of revenues for state and local governments, these sources broadly fall into two categories: recurring revenues and one-time revenues. Which are windfalls?

Recurring revenues largely come from taxes, fees, and revenue sharing, while one-time revenues primarily take the form of capital generated from municipal bond issuances. While both represent income for governments, good financial management requires that they be considered differently.

In brief, recurring revenues are used to pay for the regular and ongoing expenses a government incurs: salaries and benefits, maintenance of public facilities, welfare transfers, et cetera.

By contrast, bonds are issued to spread the cost of a capital project over its expected life cycle. This serves several functions, not the least of which being inter-generational synchronization of costs and benefits. Bonding to pay for ongoing expenses is something many governments have done, for instance to create the appearance of balanced budgets, but it’s a terrible idea and one that voters and markets eventually notice.

To be sure, what constitutes a windfall isn’t always obvious. For instance, the spike in revenues in North Dakota from the Bakken shale boom (which is largely, and wisely, being put into a permanent fund), isn’t really a windfall since it represents a secular growth in revenues that will last many years or decades (though not forever).

The mere fact that recurring income streams can be volatile doesn’t make them windfalls. Revenue volatility can make planning and budgeting difficult (which is one reason that relying on a highly progressive income tax may be a bad idea). But just as household income may be volatile, winning Powerball or receiving a large inheritance is something very different than the volatility associated with the number of hours a construction worker will work in the coming months, or the size of a manager’s annual bonus.

The existence of volatility is known in advance, even if its extent is unknown or policymakers choose to ignore its consequences. For instance, Louisiana policymakers know that volatility in oil prices impacts state revenues year-to-year. By contrast, they do not know when an oil rig is going to blow up 50 miles off their coastline leading to record Clean Water Act fines which Congress will mostly divert to states.

Windfalls, then, are one-time transfers from exogenous sources that are not designed as countercyclical stabilizers (such as an increased federal share in unemployment insurance).

So windfalls seem at first glance like they’re neither bonds nor recurring revenues. Like bonds, they are one-time income, but they don’t have to be repaid by future taxpayers. And like recurring revenues, they’re not tied explicitly to capital projects–but they’re also not recurring.

If they’re neither fish nor fowl, how should we think about managing windfalls? It’s helpful to consider the logic of how and why we issue bonds and the public choice ramifications of windfalls.

One way to think about windfalls is as bonds issued on the most favorable possible terms: zero percent interest and annual payments of nothing.

Consider a municipal bond that is floated to build a school. A county might float a $100 million bond repayable over 30 years that costs $160 million to repay in current dollars. This could make sense, as the useful lifespan of the building should be 30 years or more, and it means that the county doesn’t have to have $100 million in the bank before breaking ground.

But a school building is little use without teachers, administrators, and supplies. There are utility bills that come due monthly, books have to be replaced every few years, and the grass needs to be cut. Then there are irregular maintenance needs such as leaky roofs and broken HVAC systems. These are all expenses that are paid by general revenues. And they should be known when the county commits to issue the bond to build the school.

So it is with windfalls. When windfall money is treated as general revenues, governments can hire new employees and give raises to existing ones, expand services, and generally raise spending–all while cutting taxes.

But at some point, the windfall dries up. Then the government is left with the choice of cutting back on all the new goodies or raising taxes. Both of these are politically unpopular, to put it mildly.

Consider the very real situation facing some of Florida’s counties. Escambia County (population 302,000) could receive as much as $190 million under the RESTORE Act; this is more than half the county’s annual budget of $365 million. Some of its neighboring counties stand to gain even more as a percentage of their annual budgets.

What can Escambia County do with this windfall? On the one hand, they could spread it across, say, five years, increasing annual spending by about ten percent or cutting taxes by the same. They could conceivably eliminate their local option sales tax, which collects about $34 million annually, for about five years.

But at year six, the funds have run dry, and they’re left with the choice of either reinstating the local option sales tax or cutting services by a commensurate amount. For instance, they could eliminate all culture and recreation activities (including libraries and parks) as well as economic environment spending (which includes housing and urban development and economic development activities); those two areas cost around $34 million a year. But obviously, this is a non-starter.

So a better way for governments to think about windfalls is as very convenient bonds. Even if the windfall comes suddenly, as will be the case with RESTORE Act funds, it should still be invested in capital assets that pay benefits over a long period of time. Of course, just as schools have ongoing costs that should be paid by recurring revenues, the projects funded by windfalls may place future claims on the public fisc. So it’s important that these costs be known upfront.

This may sound familiar to people who have followed the Medicaid expansion debate. Aside from the fact that Medicaid isn’t associated with many improved health outcomes, conservatives have argued that expanding Medicaid rolls paid for by “temporary” federal dollars would later require states to either raise taxes or cut people from Medicaid rolls.

Using windfall money for projects that might otherwise have resulted in a bonded debt issuance can reduce tax liability over the long term or free up recurring revenues that might have gone to debt service. Either way, treating windfall revenues as an opportunity for capital investment rather than ratcheting up non-capital spending is a much better deal for taxpayers and those who rely on government services.

This case is consequentialist in nature; there’s no philosophical or economic rule that says that windfalls couldn’t, for instance, be used to eliminate property taxes for a decade with the knowledge they will return to ex ante levels after the windfall dries up.

Rather, the case for thinking of windfalls as bonds is about addressing the public choice issues associated with managing windfalls, which create terrible incentives for politicians and can increase volatility in revenues and spending.

Just as big capital influxes from bonds come in large amounts and irregularly, so it is the same for windfalls. They shouldn’t be treated as recurring revenues for the simple reason that they’re not recurring.

*  *  *

Daniel M. Rothschild is director of state projects and a senior fellow with the R Street Institute. He joined R Street in October 2013 having previously worked at the American Enterprise Institute and the Mercatus Center at George Mason University. His popular writing and articles and reviews have appeared in the Wall Street Journal, Reason, the Weekly Standard, the Chicago Policy Review, Economic Affairs and many other popular and policy publications. He was a 2012-13 National Review Institute Washington fellow. Dan has testified before the U.S. Congress and several state legislatures on tax and fiscal policy, government reform and disaster recovery policy.