TEXT FOR DELIVERY: 9:30 A.M., E.D.T. FRIDAY, MAY 5, 1995 ___________________________________________________________ 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 before the Joint Economic Committee UNITED STATES CONGRESS May 5, 1995 Mr. Chairman and Members of the Committee: I appreciate this opportunity to comment on the labor market data released this morning. Payroll employment was essentially unchanged in April, at 115.8 million, and the unemployment rate rose three- tenths of a percentage point to 5.8 percent. Although our published estimate of seasonally adjusted payroll employment was little changed over the month, there were some special circumstances that need to be considered to set this lack of growth into proper context. The period between our March and April surveys is always a time of employment buildup in certain seasonal industries, such as construction, services, and retail trade. Our seasonal -2- adjustment process is designed to remove such seasonal growth and allow us to discern the underlying trend in employment. It does so by examining past seasonal movements. As we discussed here last month, the growth of employment in March was boosted by mild weather, which contributed to an earlier than normal seasonal buildup both in construction and in amusements and recreation. Some of the job gains recorded in March, therefore, likely reflected growth that otherwise would have been recorded in April. In addition, our seasonal expectation for this April is much higher than it has been in recent years. One reason is that in each of the past 3 years, that is, 1992-94, there were 5 weeks between our March and April surveys, and those years were given the predominant weight in the construction of our current seasonal adjustment factors. This year, there were only 4 weeks between the March and April surveys and, thus, one less week of seasonal hiring. Partly as a result, the seasonal employment buildup expected for this April probably was too large. We have made some very rough attempts to quantify the impact of the unusual calendar pattern on the seasonal expectation of March to April employment change, although, as will become apparent, there is a large range of uncertainty connected with our estimates. One approach was to examine our historical data to assess employment growth -3- from March to April, differentiating between 4- and 5-week survey intervals. Using several techniques, we found that we might expect about 70,000 less growth in years with 4 weeks between surveys than in years with 5-week intervals. Another approach was to assume that the 841,000 seasonal growth expected for April would have occurred evenly over a 5-week survey period. This would suggest that a one-week shortfall might have depressed our seasonally adjusted employment estimate by about 168,000 jobs. Still another approach was to apply the seasonal factors for 1992 to the 1995 data. The year 1992 was the last time that the seasonal factors were based predominantly on March and April surveys separated by 4 weeks. Applying the 1992 factors to the 1995 data yields an expected seasonal increase in employment of 607,000 -- 234,000 less than the expectation implied by the actual 1995 factors. Using the old seasonal factors, then, we have a high-side estimated shortfall of 234,000 in the seasonally adjusted March-April employment changes we are reporting. Although it does seem clear that there was a "calendar effect" in April's data, given the diversity of the estimates I have just described, I cannot make a precise statement concerning its magnitude. Even allowing for the possible impact of imperfect seasonal adjustment, however, the trend in payroll employment growth clearly is down from last year's rapid pace. -4- A noteworthy development in this regard is the emerging slowdown in manufacturing (an industry relatively immune from the seasonal issues we have been discussing). Employment in the nation's factories had been on a steady upward trend since the fall of 1993, with growth picking up substantially between September of last year and this February. Over the last 2 months, however, factory employment has fallen by 39,000. In addition, manufacturing weekly hours and overtime appear to be edging down from the extraordinarily high levels of recent months, although the decline for April may have been overstated due to imperfect adjustment for the Easter and Passover observances that occurred during the reference period. Of the manufacturing industries that have been adding jobs recently, only industrial machinery continued to grow in April. Industries tied to home building and buying, such as lumber and wood products and furniture and fixtures, experienced losses over the month. Turning to the data from our survey of households, the unemployment rate increased to 5.8 percent. Although still well below the level at the beginning of 1994 -- 6.7 percent -- the rate is now higher than the 5.4 percent figure attained in December and February. Over the month, jobless rates edged up for all the major worker groups, except Hispanics. The increase in unemployment in April was -5- concentrated among reentrants to the labor force and those on temporary layoffs from jobs. In summary, the special factors affecting the April figures notwithstanding, our data suggest that payroll employment growth has slowed from the strong pace set last year. Manufacturing is showing signs of weakness, and unemployment has increased from its recent lows. My colleagues and I now would be glad to answer any questions you might have.