BY ANNOUNCING its commitment to a $4.3 billion "jobs" program of vague specification, the White House has put Congress in a bind. Congressional leaders of both parties agree that swift action is needed--preferably by adding a jobs package to a March emergency appropriations bill needed to keep unemployment benefits flowing. But they are also aware that the limitations the White House has put on its program pretty much ensure that it won't do much to help the long-term unemployed.

The president's chief economic adviser, Martin Feldstein, described the program to a Senate subcommittee last week as an acceleration of planned spending on construction and maintenance, with no special targeting on the long-term unemployed. That probably means work on highways, dams, federal buildings, mass transit, port dredging, veterans' hospitals and urban renewal.

That will be good news for the construction unions in areas where these projects happen to be located. Unfortunately, most of the long-term unemployed aren't construction workers, and certainly no one would want to attract more workers into this perennially overcrowded field.

Mr. Feldstein defends this approach in more traditionally Keynesian terms. If the projects are going to be built anyway, they might as well be built when the economy is slack for whatever general stimulus value they may have. Looked at this way, however, the money is hardly worth spending.

It is far from clear that more of this sort of general stimulus is needed. Large tax rebates are due in April as the result of overwithholding last year; another tax cut comes in July; the $5 billion in added highway spending will begin shortly, and defense spending is accelerating. If there is any punch left in deficit spending, these outlays should provide it.

Mr. Feldstein should also ask himself whether it is really likely that this "speedup" of public works won't add permanently to the deficit. When Congress looks in the pork barrel next year and finds it empty, the temptation to replace the missing $4.3 billion will be irresistible.

Studies of local public works programs have also shown that their job-creation value--even when both direct and multiplier effects are taken into account--is very low initially. In the first two years of operation, it would cost almost $50,000 a year for each direct and indirect job created in this way. The $4.3 billion would add only about 90,000 jobs a year --a drop in the unemployment bucket.

The only good reason to spend this kind of money on a jobs program is to relieve the hardships of people who have been out of work for a long time and have scant hope of being rehired. Almost 24 percent of the unemployed have been jobless for over half a year--a considerably higher number than in any other postwar recession. Their number, moreover, continues to grow. Spending $4 billion on helping the worst-off among this group could do much to get them back on their feet. If Congress and the president are interested in more than a symbolic jobs program, this is where their efforts should be concentrated.