Give everyone a job.

On January 30, The New York Times reported that President Clinton's welfare reform plan might require "a work program of 2.3 million jobs." The 2.3 million figure was actually an estimate of the number of welfare recipients who would use up the two years of cash benefits Clinton wants to offer before imposing a work requirement. According to the Times, the statistic was leaked by "an official who opposes the work program." The disclosure seemed to have the desired effect. The next day, Mary Jo Bane, co-chair of Clinton's welfare task force, assured a meeting of governors that the White House had no desire to "face the prospect none of us wants to face, which is a large number of public sector jobs."

There aren't many vocal defenders of public jobs programs in Washington these days. Conservatives who used to demand that welfare recipients be put to work sweeping the streets are now recoiling from the prospect of spending the money necessary to pay for the brooms, supervisors, and day care such "workfare" programs require. The neoliberals at the Democratic Leadership Council tend to agree that the idea of a big public works jobs plan is just too "Old Democrat." But Old Democrats don't like the idea much either, mainly because the government employee unions that support them fear public works workers might steal union jobs.

Here is the prototypical good idea that is too simple for Washington's advanced political culture to accept. There are people who need work. There is work that needs doing. Why not pay the people to do the work? It beats paying them to do nothing, which is what our welfare system mainly does, with disastrous social consequences. Yet Washington is now filled with policy planners (some of them on Clinton's task force) who will explain to you that a last resort public works program is... well, it just won't work. It's all very complicated, you see. You really have to be an expert.

But it's not that complicated. Using the government as employer of last resort is, after all, something that America did, and did reasonably well, during the Depression, when Franklin Roosevelt's Works Progress Administration employed some 3.3 million people at mostly useful tasks. In eight years, the WPA constructed 40,000 buildings, ineluding 8,000 schools. It built or improved 650,000 miles of roads, 124,000 bridges and viaducts, 8,000 parks, 18,000 playgrounds, and 2,000 swimming pools. It built New York's La Guardia Airport. It built the gym at my high school. Even some modern conservatives look back on the program with nostalgia. As one put it: "WPA--some people have called it boondoggle and everything else--but having lived through that era and seen it--no, it was probably one of the social programs that was most practical in those New Deal days." But then Ronald Reagan was always considered a bit naive.

In fact, the WPA is still a practical idea. It solves two large social problems at once. First, it is a way to keep people employed during recessions--a "countercyclical" program, in economist-talk. Second, it can be a permanent replacement for welfare programs that currently dole out cash to the poor. This second, anti-welfare function was an integral part of FDR's original conception of the WPA. When he announced the program in 1935, he simultaneously ended a relief program that had been giving cash to the ablebodied poor. "To dole out relief in this way is to administer a narcotic," FDR declared. Three years later, his aide, Harry Hopkins, would testify that

It is my conviction, and one of the

strongest convictions I hold, that the fed-

eral government should never return to a

direct relief program. It is degrading to

the individual; it destroys morale and

self respect; it results in no increase in

the wealth of the community; it tends to

destroy the ability of the individual to

perform useful work in the future; and it

tends to establish a permanent body of

dependents.

How much of the current welfare mess would we have avoided if we'd taken Hopkins' advice?

The WPA experience offers at least three other lessons that Clinton's welfare reformers might take to heart.

First, if you are running a last-resort jobs program, set the wage low. Roosevelt said the WPA's wages should be "larger than the amount now received as a relief dole, but at the same time not so large as to encourage the rejection of opportunities for private employment." The problem with implementing this sound principle today is that the amount now received as a "dole"--welfare, food stamps, and Medicaid benefits--often exceeds what can be earned in a minimum wage job. The solution, which Clinton to his credit is pursuing vigorously, is to give tax credits and health insurance to minimum wage private sector workers-and, eventually, to raise the minimum wage.

Second, get ready for a fight with the unions. The American Federation of Labor almost crippled the original WPA by demanding that it pay union wages. FDR finally had to break a strike over the issue in 1939. Today, Democrats working on Clinton's welfare plan express confidence that any differences with government employee unions can be negotiated away. That will be possible, I suspect, only if Clinton either pays his "community service" workers too much or makes sure they aren't doing anything useful enough to threaten union jobs.

Third, don't "target" public service jobs to distressed inner city communities. My high school gym--the one built by the WPA--was in Beverly Hills, California. You can see it in the movie It's A Wonderful Life (it houses the pool that Jimmy Stewart and Donna Reed fall into). At first I thought it absurd that the WPA built a gym for the youth of Beverly Hills. Now I think it represents a sound principle: Public works projects should be built wherever there are projects that need building. Many of those projects will be in the neighborhoods where poor people live, but many will be in other neighborhoods. One of the good things a neo-WPA could do is take people out of the culture of the ghetto and into the mainstream working life of the surrounding city.

It's easy to get too sentimental about the WPA. A friend of mine, a veteran California politico, once heard me getting all worked up about Roosevelt's program and decided to tell me about the WPA job he had held. It consisted of acting as bookie for a work crew assigned to paint a bridge. Even the most productive of the public works operations were not as efficient as private firms.

They never will be. They'll be even less productive in today's context, when those who would take last-resort jobs are, to a large extent, unskilled men and women with little work experience. But the jobs don't have to be very productive to be a better use of taxpayer's money than our current "income maintenance" programs. At least neo-WPA workers would be producing something.

The proper analogy, I think, is with the military. It's no coincidence that the best WPA administrators were borrowed from the Army. The military, after all, is relatively inefficient. Like a public works program, it's a big bureaucracy. It isn't subject to a lot of competitive market forces. Yet when we need it to fight, it's usually gotten the job done. And we tolerate its inefficiency because there is no better alternative. You can't fight a war with private enterprise. And, as I suspect President Clinton will discover, you can't "end welfare as we know it" by relying on private enterprise either.