This was a good employment report, but not quite the blowout that some are taking it to be. Gains were concentrated in only a few sectors, and some areas showed surprising weakness. But the labor market nonetheless remains at the tightest levels it's been at in seven years.
• Employers added 180,000 jobs in March, and January and February were each revised up by 16,000. It still looks like the February figure, now +113,000, was depressed by weather, suggesting that March's higher-than-expected total got a compensatory meteorological lift. The average of the two is just under 150,000, which feels like the current trend. That is still above the rate of population growth; to match pop growth, monthly payroll gains should be around 140,000.
• These gains were rather narrowly distributed, however. The old standbys, health care and bars and restaurants, together contributed more than a quarter of the gains. Nearly a third the total gain came from construction - but most of that from specialty trade contractors, who mostly finish buildings rather than start them. Residential building was off by 1,000, and February's loss of 3,000 was revised down to 4,000. Much of the gains in specialty trades were reversals of February's losses; residential specialty trades are still 9,000 below January levels. Manufacturing was off 16,000. Retail, however, added 36,000 (and was the source of about two-thirds of the upward revision to February employment); this is a burst of strength from what had been a lagging sector.
• Professional and business services, the heart of the post- industrial economy, fell 7,000, its first decline since November 2004, and a sharp contrast with its average gain of 37,000 in the previous six months. Declines in this sector are rare outside recessions; this is one of the things that keep us from reading this as a truly strong report. Within the sector, temp employment fell by 1,000, the second consecutive monthly decline; while February's decline could be attributed to weather, continuing weakness in this sector in March suggests not a lot of pent-up hiring demand for future months. And information, another post-industrial star, was off 5,000; finance was unchanged.
• Diffusion indexes confirmed the narrowness of the gains, with the one- and three-month measures actually falling, an unusual contradiction of the headline payroll number's month-to-month expansion. The six- and twelve-month indexes were up, but are still below their year-ago levels, and remain below historical expansion averages.
• The workweek rose 0.1 hour to 33.9 (and February's was revised up 0.1 hour); manufacturing was up 0.2 hours to 41.1, the highest in several months, but still below last summer's levels. Aggregate hours turned in a strong performance, up 0.6%, led by goods production, up 1.4%. These are some of the best numbers of the last few years.
• Average hourly earnings were up 0.3%; the back months were unrevised. Year-to-year gains fell by 0.1 points to 4.0%, their lowest level since last fall. Oddly, manufacturing pulled up the averages; earnings in private services were up just 0.2%, and yearly gains fell by 0.2 points to 4.2%. Of course, whether these numbers are still "too high" depends on what's going on with productivity. Weekly earnings were up a healthy 0.6% - not the highest level of the past year by any means, but some help to demand as mortgage equity withdrawal slows.
• Household employment, adjusted to match the payroll concept, rose 420,000 - but it fell hard in February, down 278,000. The two-month gain, then, is 142,000 - and that's not an average, that's the two- month total. No doubt weather was the culprit in the volatility, but the average gain of 71,000 for the two months is not suggesting underlying strength that the establishment survey is missing.
• The unemployment rate fell to 4.4%, matching its low for the cycle (set last October - though it's slightly lower if you want to take it out to three decimal places). The employment/population ratio rose 0.1 to 63.3% - not the high for the cycle, which was December's 63.4%, but extending the sustained strength in this measure first established last fall.
• The pool of available workers - the officially unemployed plus those classed as not in the labor force but wanting a job now - fell to 4.9% of the civilian population, matching its December level, and just 20 basis points above its all-time low, set several times in 2000.
So while this month's report isn't as mighty as it looks at first glance, it is consistent with a gradually tightening labor market. Inflation hawks on the FOMC would doubtless prefer a gradually loosening labor market, a goal that will require monthly employment gains closer to 100,000 than 150,000.
- Philippa Dunne & Doug Henwood