Yesterday’s market drop had the odor of an all-too-familiar panic about it. Once again, it feels as if the fragile fabric of policy stability in the Bush administration is threatening to unravel — as if we’re headed toward another “constitutional crisis” of the type that has driven three market bottoms over the last three quarters. Our sense of it is that, while the issues in play this time around are complex and challenging, they probably won’t develop into a full-fledged crisis. The market may very well be over-reacting.
The proximate cause of yesterday’s drop was the dollar. We continue to believe that the apparent weakness in the US dollar on forex markets is (so far, at least) a constructive unwinding of what had been excessive strength driven by years of deflationary Federal Reserve monetary policy. The press has been eager to characterize the dollar’s weakness as the result of a new policy of the Bush administration, and the atmosphere of uncertainty was made worse by what appears to be the Wall Street Journal’s deliberate campaign to embarrass and discredit Treasury secretary Snow. But the fact is that in a world of floating exchange rates, the forex value of the dollar is almost entirely determined by the Fed, not the Treasury Department (no matter how badly its secretary handles a hostile press). And the dollar’s weakness can hardly be the result of anybody’s new policy, since it’s been in decline since mid-year 2001, when the bottom in the dollar price of gold first hinted that the Fed’s deflationary siege was coming to an end.
At the same time, the market has had to deal with the disturbing spectacle of President Bush’s potentially revolutionary tax cut proposals walking a trail of tears in the Senate. And now House and Senate versions that differ both in size and structure will have to somehow be reconciled. It won’t be easy, but there were indications late yesterday that the White House has brokered a compromise between Senate and House versions that may avoid what may otherwise have been a deadlock over relatively trivial details. Whatever the precise outcome now, it seems that pro-growth tax cuts are virtually in the bag.
At this point it is impossible to predict exactly what kind of dividend tax relief can be expected, and how much. Under the Senate version, there would be 50% dividend tax elimination in 2003, 100% in 2004, 2005 and 2005, and then an automatic “sunset” that would reinstate full taxability in 2007. Obviously the bet is that the tax elimination will be extended or made permanent during what is presumed to be a second Bush administration. The immediate impact on the stock market from passage of a bill structured this way would depend critically on the market’s assessment of the likelihood of extension or non-extension. For example, we have calculated that full, permanent dividend tax elimination would result in an immediate windfall to stock prices of about 15% across the board; under the Senate version, that 15% would be approximately linearly reduced by the probability of non-extension, so a 50% probability would result in an 8% windfall. We would expect to see dividend tax elimination drive long-term changes in corporate behavior, such as increased pay-out ratios and greater reliance on equity financing. But such changes would not necessarily be linear with the probability of non-extension, perhaps requiring very high levels of long-term certainty.
The Senate version differs from Bush’s original proposal in two other important respects, one good and one bad. The good one is that the Senate version does away with the requirement that tax-free dividends be paid only from corporate profits that have been fully taxed going forward, after passage of the new law. This would permit companies with large cash hoards, such as Microsoft, to pay them out tax-free. The bad one is that the Senate version does away with the basis step-up for capital gains taxes that would effectively eliminate the double-taxation of retained earnings. Without the step-up, the Senate’s version puts in place an arbitrary incentive at the margin for companies to pay out earnings as dividends rather than to retain them for future re-investment in the business.
The House version differs from Bush’s original version in yet other ways. First, it doesn’t eliminate the taxation of dividends, but instead caps it at 15%. The 15% cap would be in place for ten years, after which it would automatically sunset, and full taxation would be reinstated unless the cap were extended or made permanent. As with the Senate version, the impact on the stock market would depend on the market’s assessment of the likelihood of extension or non-extension. Again assuming a 50% probability of non-extension, the immediate windfall for the stock market would be 4.4% under the House version, compared to 8% for the Senate version. If we assume that the long ten year life of the 15% cap reduces the probability of non-extension to only 25%, then the windfall rises to 6%.
The House version, like the Senate version, does away with Bush’s proposed basis step-up for retained earnings. But, importantly, it puts in its place a reduction in the top capital gains tax rate from 20% to 15%. It is difficult to model the windfall or long-term effects of such a reduction, but in our economic model they would likely be quite considerable. Remember, capital gains beyond those explained by retained earnings can only result from unexpected increases in earnings growth rates — so by reducing the capital gains tax rate, a specific incentive for entrepreneurial risk-taking is established. Even without specific modeling, we are confident that the proposed cut in capital gains taxes in the House version would more than compensate for the House’s lower level of dividend tax relief compared to the Senate version.
Which version or combination will prevail is not just a question of preferred features — it’s a matter of budgetary impact, as measured under Congressional rules that treat all tax cuts as deadweight costs. Measured that way, the dividend relief in the Senate version has a cost of $124 billion, while the cost of the dividend and capital gains relief in the House version is $280 billion. With what looks now like a total likely tax cut of only about $350 billion (the largest cost likely to be approved by hold-outs in the Senate), the only way the House version can survive would be to narrow down its sunset considerably from the current ten years.
The market is no doubt discomfited by the gimmickry of sunsetting, and all the rest of the legislative sausage-making it has had to witness since Bush first proposed these tax cuts on January 7. But the fact is that such things are necessary in what is, inescapably, a structurally deadlocked legislative environment. We suspect that when this is all over, the market will step back, take a deep breath, and begin to appreciate just how much the Bush administration was able to achieve against all odds. It will by no means be all we could have hoped for. But it will be far more than most will have expected.
