OVER THE last three decades, the proportion of black youths holding jobs has dropped dramatically. Researchers have long argued whether to blame the labor market or the youths themselves. A new study provides dramatic evidence that one factor--the unwillingness of black teen-agers to work at low-paid jobs--can be ruled out as an explanation. It also suggests that the gap between black and white youth employment could be closed at a surprisingly low cost.

In 1955, more than 50 percent of all black males aged 16 to 19 held jobs--roughly the same proportion as for their white counterparts. In 1981, the white male youth employment rates had stayed about the same, but the rate for black male youths had dropped to about half that. Now, with unemployment still higher, only 16 percent of black teen-agers hold jobs as compared with 40 percent of white teen-agers.

Alarmed by the continuing drop in black youth employment, the Carter administration launched several large-scale experimental projects that succeeded, temporarily, in reversing the trend. The largest was a three- year project that guaranteed low-wage jobs to youths who stayed in school. The project, which operated in 17 low-income areas around the country, was intended to improve the future job prospects of its participants by encouraging them to remain in or return to school and by giving them practical job experience.

The early findings are encouraging. Dropout rates declined somewhat in the project sites, and a significant number of dropouts were persuaded to return to school--although those returning were far more willing to enroll in alternative education programs than to go back to the schools they had been attending. But the most dramatic effect was that students were so anxious to take the part-time jobs offered--and project directors were so successful in finding appropriate jobs-- that the project raised employment rates for its predominantly black, low-income constituency to the point where they were close to the comparable rates for middle-class white teen-agers.

The projects had little success in inducing private employers to participate. Even with the promise of full payment of the youths' wages, only 18 percent of employers contacted agreed to help out, and these were mostly small retail stores. With lower levels of subsidy, private employer participation fell off rapidly. But public and nonprofit agencies provided useful jobs for more than three-quarters of the 76,000 students who participated.

The Manpower Development Research Corporation, which evaluated the project, estimates that it could be done in all poverty areas of the country at an annual cost of about $600 million and in all areas for about $1.7 billion. That's not a large price to pay for eliminating one of the most pernicious and enduring of the nation's social problems.