Excluding some dividends and all economics

Posted: May 15, 2003 12:00 AM

I recently worried that the Senate Finance Committee's efforts to slice up a low-calorie tax snack would turn into a stew with too many turnips and not enough beef. That is exactly what happened. Instead of eliminating or reducing the individual tax rate on dividends, the committee opted for a zero tax on the first $500 of annual dividends -- taxing all but 10 percent of any extra dividends at full income tax rates. This antique was reportedly dusted off by Olympia Snowe, Republican from Maine.

Before launching this attempted policy coup, CQ Today noted that the committee booted all economists from the room. Easy to see why. Mixing middling tax exclusions with steep tax rates is the textbook definition of foolish tax policy -- it keeps high marginal tax rates on a thin tax base. And that minimizes economic benefits while maximizing revenue lost: no bang for the buck.

The Economist figures the Grassley-Snowe scheme reduces individual's half of dividend tax collections by merely a fifth. Yet Sen. Grassley boasts that 86 percent of investors would be exempted from the tax. That means the other 14 percent of investors have been paying 80 percent of this tax, and Grassley's committee now wants the select few to pay 100 percent. Good luck. There's a sucker born every minute, as honorary economist P.T. Barnum taught, but more turn out to be mediocre senators than good investors.

In Washington, sticking 14 percent of taxpayers with 80 percent to 100 percent of any tax is described as "fairness." And cutting any tax is described as "regressive." Brookings Institution scholars Bill Gale and Peter Orszag, for example, condemn the Senate Finance Committee bill as "regressive" because "more than one third of filing units would receive no benefit." They receive no benefit for the same reason they are called "filing units" -- they are not taxpayers. That is also why opinion polls find that more than one third of these filing units don't give a hoot about tax cuts: They are not taxpayers, and few are voters.

The twin notions that all tax cuts are "regressive" and that having a few people pay for government is "fair" can be statistically massaged into a "distribution table." Such tables purport to show how much of any tax cut goes to taxpayers at various income levels. But the only way to do that is to start with the incredible assumption that people will hold the same shares of dividend-paying stocks in taxable accounts if dividends are taxed at 15 percent to 20 percent as they did when dividends were taxed at 39.6 percent.

The same nonsensical assumption explains why the Joint Committee on Taxation imagines that cutting the dividend tax to 20 percent might lose almost as much revenue (about $9 billion a year) as the Grassley-Snowe gesture.

In reality, a 15 percent to 20 percent dividend tax is likely to cost the government nothing because the Treasury would quickly find itself with many more dividends to tax. If nobody asks revenue estimators the right questions, nobody gets the right answers.

Distribution tables based on who reported dividend income when dividends were taxed at rates up to 39.6 percent tell us nothing about who will report which sorts of income after the dividend tax rate falls. Many investors in middle and upper tax brackets would surely report much more dividend income. But they would likewise report less from capital gains and interest income, which are not on that same distribution table.

To create distribution tables pretending to show what might happen to just dividends alone -- by assuming that the entire mix of taxable and tax-exempt assets is unaffected by tax rates -- is to engage in a bookkeeping fraud of Enron-like proportions.

Grassley's gullibility about such tables was revealed in his prideful remark that 86 percent of investors would pay no dividend tax under a $500 exclusion. But that is precisely because 86 percent of investors have heretofore carefully avoided holding overtaxed dividend-paying stocks except in tax-exempt pension funds.

If 86 percent of investors choose to collect less than $500 in dividends, that just proves the dividend tax is prohibitively high. What the Senate Finance plan says to investors -- loud and clear -- is you would be foolish to put too much into taxable securities that pay dividends. Once you reach the $500 dividend limit (an investment of only $25,000 with a 2 percent yield), it is then time to sell some stocks put any extra savings into pension accounts or tax-exempt bonds. This scheme also tells more serious investors to continue favoring companies that retain earnings over those that pay dividends. And that tells corporations like Enron and WorldCom to skip dividends and instead use spare cash and risky debts to make bad investments in snatching-up companies they cannot prudently manage.

When it comes to dividend taxes, the Senate Finance Committee tried to divert the Senate away from two fast freeways, one marked House and the other White House. Instead, it put out a map pointing down a narrow one-way street with a dead end.

Even with nonsensical revenue estimates that assume investors never choose to invest more in taxable stocks regardless of taxes, there are several options that look just as parsimonious as the Senate Finance scheme. The easiest is to simply cap the dividend tax at 20 percent for 10 years. If revenue estimators dare claim that plan costs one dollar more than the Senate Finance Committee's, they should all be fired for incompetence. It would be nicer still to take the tax rate down to 15 percent to 17 percent, but then it would be for fewer years so its durability might be less secure.

Another option is to take the dividend tax down to zero from 2004 to 2007. Such a 3-year sunset is like trying something out to see how it works. That is not necessarily foolish, contrary to New York Times columnist Paul Krugman. What was truly foolish was an earlier pie-in-the-sky plan from Grassley's committee to gradually reduce the tax on dividends to zero in 2005 only to bump it back up to 35 percent the following year. Investors would then want companies to delay paying dividends until the one year in which dividends caught the big break. The confidence of investors and employers (which necessarily precedes confidence of consumers) will never be cheered up by such distant promises, but only by something tangible that happens soon.

Among several potentially marvelous ways of dealing with dividend taxation, the Senate Finance Committee's antiquated gesture of excluding a portion of dividends and then overtaxing the rest holds no promise at all. It deserves a veto. Like many economic policies created by banishing economists from the room, this one won't last long.