What President Bush's three tax-deferred savings proposals offer -- and don't offer -- American consumers.
New York Times; New York, N.Y.; Mar 13, 2003; Hal R. Varian

A TAX economist I know once told me about his March ritual. He would pick a quiet Sunday, sharpen his pencils, set out a plate of tea and cookies, and immerse himself into that most delightful of pastimes, doing his taxes.

To him, every line of the tax forms had a story: some stories were noble, some venial, but all were fascinating. What a treat it was for him to participate in this vast and thrilling drama.

To the rest of us, tax preparation holds considerably less appeal.

In fact, to most Americans it is the most unwelcome of rituals, full of confusion and pain.

President Bush's proposed tax plan would no doubt add to the confusion, though it might reduce some of the pain, at least for those with high incomes.

The two main pillars of the Bush plan, eliminating taxes on some dividends and extending tax-exempt savings programs, each has some merits. But if either is actually enacted, we will need a whole new set of epicycles within epicycles to incorporate them into the already unwieldy tax system.

As many people have remarked, most benefits of these plans accrue to the wealthy, but this is true for almost any form of tax reduction, since the wealthy pay most of the taxes.

A more damning sin, to many economists, is that the Bush plan is a budget breaker. There may be some compelling economic logic to cutting taxes on dividends, or subsidizing savings, but this should be done in a fiscally responsible manner -- by identifying a specific revenue cut or tax increase to balance the books.

The proposals for the three new tax-deferred savings plans are a notable case in point. Encouraging savings is a worthwhile goal, as is tax simplification. The proposed Lifetime Savings Accounts, Retirement Savings Accounts and Employer Retirement Savings Accounts consolidate and simplify several existing programs. This is a good thing, but the annual contribution limits are far too generous. Unifying tax-deferred retirement programs, but keeping the current limits on contributions, would be a more prudent policy.

Under the first two plans you contribute after-tax dollars, but pay no subsequent taxes on earnings and withdrawals, as with today's Roth I.R.A. plan. With the Employer Retirement Savings Plan you contribute before-tax dollars, but then pay taxes on the withdrawals, as with current 401(k) plans.

Investment earnings are tax-free under all three programs. This is obvious for the first two, since you put after-tax money in, and pay no taxes on the money withdrawn. It's not so obvious for the employer accounts, so let's look at the arithmetic.

Suppose you are a year from retirement, you are in a 40 percent marginal tax bracket now, and expect to be in that tax bracket after you retire. You have an investment opportunity that can earn 10 percent, and have $10,000 of before-tax earnings from your employer that you want to save.

If you use a Lifetime Savings Account or Retirement Savings Account, you first pay taxes on the $10,000, leaving you with $6,000. You then invest this money and earn 10 percent, leaving you $6,600 after one year, with no additional taxes due.

Alternatively, you can have your employer invest the $10,000 of pretax money in the employer account, which grows to $11,000 next year. When you withdraw this money you have to pay $4,400 to the I.R.S., leaving you with $6,600, just as before. Either way, the government collects $4,000 in present value as tax, and you escape all taxes on the 10 percent interest payment.

Any of these plans is a good deal compared with investing outside a tax-sheltered account. In that case you would have to pay taxes on both the $10,000 income and the $600 of interest, leaving you with only $6,360 to spend.

The proposed savings plans would be a great deal for those with the money and inclination to save. Unfortunately, these plans are unlikely to create much new savings. Instead, those with money to save would simply shift it from existing taxable accounts to tax-sheltered ones. Those with low incomes would probably not find the tax incentives enough of an inducement to increase their savings rate.

So how can we increase the savings of low-income workers? A recent National Bureau of Economic Research working paper offers some interesting analysis. (''For Better or for Worse: Default Effects and 401(k) Savings Behavior,'' by James Choi, David Laibson, Brigitte Madrian and Andrew Metrick. The first two authors are from Harvard; the latter two are from Chicago and Wharton, respectively.)

The economists looked at data from three employers that offered automatic enrollment in 401(k) plans. Employees could opt out, but they had to make an explicit choice to do so.

They found that the participation rate in these programs was spectacularly high, with over 85 percent of workers choosing the default option and enrolling in the 401(k) plans.

Unfortunately, almost all these workers also chose the default investment, typically a money market fund with very low returns and a low monthly contribution. (Presumably the employers made the default investment highly conservative to eliminate downside risk and possible employee lawsuits.)

The net result was a positive, but quite modest, increase in overall savings.

In subsequent work, the economists examined the experience at a company where there was no default: within a month of starting work, employees were required to choose either to enroll in the 401(k) plan or to postpone enrollment.

By eliminating the standard defaults of nonenrollment, and of enrollment at low rates, this ''active decision'' approach raised participation rates from 35 percent to 70 percent for newly hired employees.

Moreover, employees who enrolled in the 401(k) plan overwhelmingly chose high savings rates.

It appears that defaults exert a powerful pull on behavior, perhaps too powerful. Forcing employees to make a choice seems to have a positive, and welcome, effect on overall savings.