TEXT FOR DELIVERY: 9:30 A.M., E.D.T. FRIDAY, JUNE 3, 1994 ___________________________________________________________ Advance copies of this statement are made available to the press under lock-up conditions with the explicit understanding that the data are embargoed until 8:30 a.m. Eastern time. ____________________________________________ Statement of Katharine G. Abraham Commissioner Bureau of Labor Statistics on June 3, 1994 Thank you for joining us here this morning. I would like to offer a few comments to supplement today's Employment Situation news release, and my colleagues and I would then be happy to answer any questions that you may have. As was reported earlier this morning, the nation's unemployment rate fell in May and the number of payroll jobs continued to rise, but at a more modest pace than in the prior 2 months. The measure of total employment from the household survey registered a large gain in May; this brings its year-to-date increase more in line with that exhibited by the payroll survey. As an aside, I would note that payroll employment estimates have been revised to reflect the benchmarking to the March 1993 universe count and -2- updated seasonal factors. I will discuss these revisions in a few moments. Since December, the number of payroll jobs has increased by 1.2 million, an average of 247,000 a month. Since January, the first month in which household estimates were obtained from the revised Current Population Survey, total employment has risen by about 900,000, an average of 225,000 a month. More than half a million of this gain was recorded in May. Results from both surveys show the pace of job growth thus far in 1994 to be somewhat above the average for 1993. The over-the-month fall in the jobless rate, from 6.4 to 6.0 percent, brings the total decline since January to seven-tenths of a percentage point. The rates for adult men and adult women have declined by roughly equal amounts since the start of the year; the rate for teenagers has shown no clear trend. Whites, blacks, and Hispanics all have shared in the unemployment rate reduction. The number of unemployed persons, 7.9 million in May, has fallen by about 800,000 since January. In examining these figures, data users should remember that, whenever unusually large movements occur in a single month, the magnitude of those changes often turns out to have been overstated once additional data become available. It nonetheless is clear that unemployment continues to trend downwards. -3- Each month since January, the Bureau has prepared a projection of what the unemployment rate would have been had we continued to conduct our household survey using pre-1994 methods. The graph attached to my statement plots our best estimates of unemployment as it was or would have been measured, using both the old and the new survey methods, for the period January 1993 to May 1994. This graph should, however, be interpreted cautiously since the projections for the later months of 1994 are inherently less reliable than those for the earlier months. Beginning next month, we had planned to replace these unemployment projections with data derived from a special small-scale parallel survey. This parallel survey using the old pencil-and-paper collection methods was initiated at a time when there were serious questions about how the redesigned household survey would perform in its early months. Data from the redesigned survey have, in fact, behaved quite reasonably and seem to have been well accepted. Given the competing pressures on the Bureau's budget, together with serious technical concerns about the quality of the data from the new parallel survey, we have decided to discontinue its collection. Returning to the results of our employer survey, the increase of 191,000 jobs reported for May followed much larger gains in March and April. The May rise included a return of about 70,000 strikers in the trucking industry; -4- absent their return, the May gain would have been about 120,000 jobs. Services accounted for most of the over-the-month job rise, net of the gains associated with the return of striking workers. But in this industry, as in many others, the pace of growth was much slower than in the prior 2 months. Business services, in particular, registered an unusually small increase. The expansion in construction employment also slowed in May. Factory employment was flat, with no noteworthy changes among the individual manufacturing industries. Both the manufacturing workweek and factory overtime hours eased back a tenth of an hour from extremely high April levels. Finance, insurance, and real estate experienced an employment decline, reflecting in part the impact of rising interest rates on refinancing and home purchases. Employment gains in retail trade were somewhat more modest than in recent months, but, generally speaking, retail employment growth has been quite strong since late last year. Today, we are issuing revisions to the payroll employment series to incorporate our routine annual benchmark adjustments. Each year at this time, we re- calibrate our sample-based survey estimates to full universe counts, derived principally from a separate BLS program that aggregates administrative records from the state -5- unemployment insurance tax system. The benchmark adjustment to the seasonally adjusted March 1993 level is an upward revision of 239,000 jobs, or two-tenths of one percent. Estimates for the post-benchmark period, April 1993 forward, also have been recalculated, in accordance with standard practice, to incorporate the shift to a new benchmark level as well as recomputed bias adjustment and seasonal adjustment factors. The benchmark process resulted in an average upward adjustment of about 25,000 jobs per month for the period since April 1992, indicating somewhat stronger job growth than originally reported. The revisions to the employment estimates for the benchmark and post- benchmark time periods are well within historical ranges. In summary, the job market continued to improve in May. The unemployment rate declined and total employment from the household survey registered a substantial increase, bringing job growth in this survey about in line with that from the payroll survey for the year thus far. Payroll job growth was more modest than it had been in March and April, but this must be viewed in the context of the past 2 months' growth having been exceptionally strong. My colleagues and I now would be glad to answer any questions you might have.