Red sails in the sunset: the "sunset" tool with the Bush team's help is subverting the U.S. budget process.
Rules governing the federal budget process have evolved continually since the passage of the Congressional Budget Act in 1974. Attempts at direct deficit control during the Gramm-Rudman-Hollings period of the late 1980s were followed by the discretionary spending caps and "PAYGO" restrictions on tax cuts and entitlement increases of the 1990s. Since 2001, the controls have been ignored. Rapid growth in discretionary spending and a series of tax cuts have contributed to exploding deficits. In place of explicit limits on the size of tax cuts, President Bush and Congress have adopted the procedure of setting a target revenue cost to be achieved over a ten-year budget "window." In 2001, the target tax reduction was $1.35 trillion. In 2003, the target (under the Jobs and Growth Tax Relief Reconciliation Act) was a more modest $350 billion, and key Senate Republicans threatened to vote against any tax cut with a larger ten-year price tag.
The unenviable task of determining the ten-year revenue cost falls to the Joint Committee on Taxation (JCT). There has been much discussion about whether JCT's methodology should include "dynamic" scoring, incorporating the macroeconomic feedback effects of tax legislation on revenues. But lack of success on the dynamic scoring front has not deterred those who seek to squeeze large tax cuts into small ten-year revenue totals. The sunset provision has become the answer. If each year's net revenue cost cannot be made smaller (through dynamic scoring), then the number of years during which the tax cut officially applies can be reduced.
Sunsets made an impressive appearance in the 2001 law, when most provisions expired at the end of December, 2010. Morbid jokes about taxpayer responses to the legislated 2011 reinstatement of the estate tax (after a full re-peal in 2010) did little to slow the further use and refinement of the technique. By 2003, Congressional tax writers had become sunset masters, varying the year of sunset as they modified provisions in order to keep the overall revenue cost magically fixed at $350 billion. In the end, the legislation scheduled some provisions to disappear after 2004, others alter 2005, and still others after 2008, all well before the 2013 end of the ten-year budget window. Thus, sunsets allowed Congress to impose true revenue losses that far exceeded stated targets. Those crafting the legislation encouraged taxpayers to view the provisions as permanent.
The rise of the sunset illustrates how legislators have subverted their own attempts at self control. Traditionally, ns in Europe today under the Stability and Growth Pact, U.S. governments have focused on revenues, spending, and the deficit for the coming fiscal year. But over the years, Congress has lengthened the relevant budget window, first to five and then to ten years. The reasoning was that a longer window would retard the use of short-term timing devices to overstate revenue with in the budget period, for example by changing the timing of tax payments due near the end of the budget period. With more years within the budget window, so the argument went, the impact of one year's timing tricks would be diluted. Unfortunately, using a multi-year budget window hasn't eliminited the incentive for revenue shifting. The Taxpayer Relief Act of 1997 included a provision permitting taxpayers to transfer funds from traditional "front-end" IRAs to "back-end" Roth IRAs, deferring some taxes due on the transferred funds for as long as three years. Although the transfer provision was a revenue loser in present value, estimates projected increased revenue within the five-year budget window because of the taxes paid on with drawals from traditional IRA accounts. President Bush's most recent budget, for fiscal year 2004, proposed similar shifts among tax-sheltered savings plans that looked good even for a ten-year budget window.
The multiyear budget window, then, has two apparently paradoxical effects on the design of revenue provisions. On the one hand, legislators still seek to shift revenue costs beyond the budget window, whether to year six for a five-year window or year eleven for a ten-year window. On the other hand, they also seek to shift revenue costs to earlier years within the budget window through the use of sunsets. Each of these responses makes sense if one considers the incentives that give rise both to budget restrictions and to attempts to circumvent them.
Restrictions on tax cuts and spending increases reflect the general consensus that, without them, we will shift too much of the burden of government onto future years and generations. Though we may agree that such burdens shouldn't be imposed on our grandchildren, we need budget rules to keep Democrats and Republicans from cheating. Otherwise, the two sides will be locked into a dynamic prisoners' dilemma. Each side, when in power, will seek to cut taxes and/or increase spending in its areas of priority simply because it rationally anticipates that the other side will do the same when it has an opportunity.
Enter the multiyear budget window, with restrictions on tax cuts and spending increases. Those in power clearly will seek to pass the buck--to make costs appear beyond the window, in a period when others may be in control. But, if the budget window is long enough, they will also prefer to spend the money sooner rather than latch After all, they may no longer be in power by the later years of the budget period and, if they are, they can renew the legislation before it expires. Shortening or lengthening the window reduces one problem but exacerbates the other.
What's the solution? The answer lies in the shape of the budget window, If the budget window ends precipitously, there will always be an incentive to shift costs into the period just beyond; if all years of the budget window count equally, then there will also be an incentive to opt for spending and tax cuts early in the budget period--to use sunsets. Altering the shape of the budget window, to include perhaps many years, but counting later years progressively less, would eviscerate both of these incentives, leaving the timing of legislated provisions to be determined by their merits, not budget politics. This modification would not be a complete solution. By itself, it leaves uncontrolled the implicit liabilities to pay future entitlement benefits like Medicare and Social Security. But it would be an improvement over the current approach.
Alan J. Auerbach is Robert D. Burch Professor of Economics and Law and Director of the Robert D. Burch Center for Tax Policy and Public Finance at the University of California, Berkeley.
|Printer friendly Cite/link Email Feedback|
|Author:||Auerbach, Alan J.|
|Publication:||The International Economy|
|Date:||Sep 22, 2003|
|Previous Article:||Revisiting Sarbanes-Oxley: was the well-intentioned landmark legislation slapped together too quickly?|
|Next Article:||King of threadneedle street: how Mervyn King reshaped the Bank of England.|