THE STUDY DEMONSTRATED THAT THE RATIO OF UNSKILLED TO SKILLED EMPLOYMENT AND UNEMPLOYMENT VARIES FROM THE PEAK TO THE TROUGH OF THE BUSINESS CYCLE. DATA PRIMARILY FROM "EMPLOYMENT AND EARNINGS AND MONTHLY REPORT ON THE LABOR FORCE," MARCH 1967 WERE USED IN RELATING THE ANNUAL UNEMPLOYMENT RATES FOR CRAFTSMEN AND LABORERS TO BUSINESS CYCLE PEAKS AND TROUGHS OVER THE PERIOD FROM 1947 TO 1966. THREE HYPOTHESES WERE INITIALLY PROPOSED TO EXPLAIN CYCLICAL VARIATIONS IN THE RATIO OF LABORERS TO CRAFTSMEN UNEMPLOYMENT--LABOR HOARDING, DOWNGRADING, AND JOB DILUTION. ALL THREE SUGGESTED THAT THE RATIO OF UNSKILLED TO SKILLED UNEMPLOYMENT WOULD RISE AS LABOR MARKET CONDITIONS LOOSENED. CONTRARY TO EXPECTATIONS, THE DATA SHOWED THAT THE RATIO FELL WHEN CONDITIONS LOOSENED. THE USE OF RATIOS SUGGESTED THAT THE PARADOXICAL BEHAVIOR OF THE RATIO OF UNSKILLED TO SKILLED UNEMPLOYMENT COULD BE EXPLAINED IN OTHER WAYS--(1) BY USING MONTH-TO-MONTH INDICATORS OF PEAK TO TROUGH VARIATIONS IN RELATIVE UNEMPLOYMENT RATHER THAN ANNUAL CHANGES, (2) BY AGGREGATING THE UNEMPLOYMENT DATA IN OTHER WAYS, AND (3) BY CONSIDERING WHAT APPEARS TO BE DIFFERENT RELATIVE EMPLOYMENT AND UNEMPLOYMENT BEHAVIOR OVER THE BUSINESS CYCLE. THIS DOCUMENT APPEARED IN "MONTHLY LABOR REVIEW," VOL. 90, NO. 8, AUGUST 1967. (ET)