Dynamic Treasury February 13, 2006; Page A16
Believe it or not, there is some good news to report from Washington. The Treasury Department wants to create a small new shop to study the impact of tax changes on economic growth and thus also on federal revenues.
President Bush's Fiscal 2007 budget last week requested about half a million dollars to establish the office of "dynamic analysis" within Treasury's tax division. The not-so-novel idea is to assess the macroeconomic impact of tax policy, with an eventual goal of more accurately measuring the consequences of both tax cuts and tax hikes on federal revenue.
If you're wondering why the feds don't already do this, join the club. These columns have been urging this for going on 30 years, and it would seem to make good policy sense since tax changes are proposed and sold in part for their economic benefits. Yet the bureaucracy, both in Congress and in most of the Treasury, calculates the impact of tax changes largely on a "static" basis, which assumes little impact on taxpayer or investor behavior. So in the crudest example, cutting income tax rates by 20% is estimated to cut revenues by roughly 20% as well, while a tax increase of that amount gets "scored" as raising a like amount of revenue.
In the real world, this almost never happens. That's because everyone outside of Beltway Land knows that tax changes have a huge impact on business and individual incentives -- on how and where to invest, how much to work and save, and whether to seek tax shelters to avoid confiscatory tax rates. You'd think politicians would want to take at least some of that into account when contemplating tax changes.
Expect to read in the coming days from liberal critics that supply-siders are hijacking Treasury to show that "tax cuts pay for themselves." But the real goal here is accurate score-keeping that takes into account the real impact that taxes have on the economy. And no supply-sider we know -- and we know them all -- claims that all tax cuts pay for themselves on a dollar-for-dollar basis.
Not all tax cuts are created economically equal because they have different effects on incentives. Some of the most politically popular are also some of the least economically useful. Tax "rebates," for example, are largely a one-time cash transfer, so they don't affect long-term incentives. They thus cost Treasury a bundle because they get very little economic bang for the buck.
Cuts in marginal income tax rates, on the other hand, have demonstrated over the years that they have a big impact on investment and work decisions and thus recoup much of the revenue that "static" models estimate they will lose. Harvard's Martin Feldstein has done research calculating that marginal rate cuts yield back at least one-third of their projected revenue loss.
He's also shown that such cuts have an especially large impact on the decisions of household second-earners, typically women, to enter the workplace. A 50% marginal tax rate on the next dollar of income -- the combined state and federal burden in such high-tax states as New York and California -- reduces greatly the incentive for a spouse to trade staying home for a 40-hour week.
Cuts in capital gains tax rates have had an even larger revenue payback over the years, because they raise the after-tax return on capital and because of the incentive they give investors to lock in a profit by selling stocks and thus creating a taxable event. The 2003 tax cuts on capital gains and dividends have been producing far more revenue than the official estimates expected, including a surprise $100 billion windfall last year. As White House Budget Director Josh Bolten told Congress last week, that's real money even in his business.
All of which is to say that the Treasury proposal is much needed and deserves support from Congress. Come to think of it, Congress should undertake a similar effort at the Joint Committee on Taxation, which has a record that makes Treasury look prescient and open-minded.
| URL for this article: http://online.wsj.com/article/SB113980143846172184.html | |
No comments:
Post a Comment