Robert J. Shapiro of the Progressive Policy Institute may feel a bit like a lecturer who, having donned his newest, most elegant suit and delivered the speech of his life, is approached by a member of the audience whose only comment is: "Nice socks."

I've just seen Shapiro's paper on the federal deficit, a thoughtful bit of maneuvering between the rocks of conservative program reduction and the shoals of liberal pressure for tax increases. The vice president of the Washington-based institute, and its director of economic studies, makes the case that large deficits of the type we now have are not progressive, both because their burden falls disproportionately on the middle class and because they "paralyze the political will and imagination of policy makers."

He assesses the reasons for the deficit -- spending or taxes? -- and also offers a way out of the mess. That's the elegant suit. Here, in a subparagraph on Page 3, are the "nice socks":

We should "divide the budget into a 'Public Consumption Account' covering government operations and transfer payments and an an 'Investment Account.' " Then we should require that the government collect enough revenues each year to cover the cost of the first account, authorizing deficit spending only for the second.

"This arrangement," he argues, "offers several advantages. By requiring the government to raise enough revenues to cover all public consumption, the approach disciplines the tendency of many politicians to cut taxes without cutting spending. Moreover, by limiting any increase in the deficit to new investment commitments, the rule channels into wealth-producing activities the related tendency to increase spending without raising taxes. And by letting the government borrow only for purposes that will improve people's economic prospects, the country's capacity to pay off the loan is enhanced as the loan itself is extended."

Under the Shapiro scheme, as he explained in an interview, spending for infrastructure, for education and training, for research and development, all would be financeable through deficit spending, because those things promise to enhance the future income stream. Spending for housing, for health care for the elderly and for income maintenance programs of various sorts would have to be paid for out of current revenues, because they do not increase future income.

The obvious objection is that proponents of various programs and entitlements would simply redefine them as "investments" and continue business as usual.

Two responses: In the first place, there might be reasonable tests for determining what is a transfer and what is an investment (though in some cases -- health-care expenditures, for instance -- the distinction might be hard to draw). But the more important point is that the very necessity of having to think in investment terms in order to justify deficit spending might lead proponents to cast their programs differently, and more effectively.

Advocates for the poor, for instance, would be under pressure to propose programs that improved the future earning prospects of the poor: programs for helping the poor out of their poverty as opposed to simply taking better care of them.

Shapiro's consumption/investment distinction would be preceded by another rule: replacement of the Gramm-Rudman deficit targets with a new rule that would allow deficits to grow, but only by the rate at which per-capita income grew in the preceding year. (The limit would be lifted in wartime and recession.)

"If you accept the view that large permanent deficits drain both the resources and the will to address major problems, then what you want to do is be sure that activities of government that cost current dollars but do not produce future dollars not be financed through deficit spending," he explained. "The argument isn't whether we should or shouldn't do something, but only how we should pay for it. It's like saying a family could reasonably borrow money for college tuitions, because that enhances future income, but not for dinner at an expensive restaurant."

He thinks his idea might provide a way out of the box that the hard right and the traditional liberals have unwittingly conspired to create.

"Hard-right conservatives accept deficits as the price of cutting taxes and also as a club against new programs. The liberal position in the '80s held that the deficit was progressive, because it was progressive 25 years ago. But then as they saw pressures coming on liberal programs from deficit, they were forced to the position of supporting higher taxes in order to oppose the conservative deficit and preserve liberal programs."

The burden of his paper, "Paying for Progress" (PPI, 316 Pennsylvania Ave. SE, Washington, D.C. 20003) is that large, persistent deficits are counter-progressive. But an interesting byproduct of his approach would be to force liberals to evaluate their proposals, not in terms of making the disadvantaged more comfortable but in terms of enhancing their future earning power.

Nice socks.