Jobs: Slim Pickings in March

Payroll growth was disappointing, which ought to keep the Fed from changing its pace. Another lesson from the numbers: Stay in school

By David Wyss

After February's impressive, 243,000-job addition to the payrolls, March pretty much killed the optimism, with a meager 110,000 increase. The unemployment rate fell to 5.2%, however, with the household and payroll data remaining out of sync. The weak payrolls number came as a surprise given the strong survey reports and the drop in initial claims for unemployment insurance -- the market consensus had predicted a 225,000-job gain (see BW Online, 4/4/05, "A March Hiccup for Employment").

The detailed information also contained disappointing numbers. Manufacturing jobs decreased by 8,000 after eking out a gain in February. Retail slid 10,000, a surprise especially given the early Easter, which usually boosts retail sales. However, Easter counts as less of a factor nowadays in retail than it used to.


  Despite the March jobs anticlimax, monthly payroll gains have averaged 166,000 over the last 18 months. This represents solid growth, although only two-thirds the monthly average experienced from 1992-2000. With an unemployment rate near 5%, economists probably should expect gains to average in that lower range over the coming months -- about the increase necessary to absorb new labor-force entrants and thus keep the unemployment rate flat.

The labor-force participation rate remains down 1.5 percentage points from its March, 2000, peak, at 65.8%. This is normal. A a one-percentage-point rise in the unemployment rate is usually associated with a one-percentage-point drop in participation, and the rate has risen from a trough of 3.8% to 5.2%.

The household employment data contain much more encouraging news. The government conducts two separate surveys of the labor market, with the household survey based on a sampling of 90,000 households, while the payroll survey derives from a poll of corporations. Household employment rose 357,000, lowering the unemployment rate to 5.2%, from 5.4% in February. The two series often diverge month-to-month, reflecting the different samples and seasonal factors.

Normally we at Standard & Poor's trust the payroll survey more for month-to-month changes, but the household survey still warrants some attention. Over the last 12 months payrolls rose 1.82 million, while the household data show a gain of 2.68 million.


  This divergence is par for the course, caused largely by a rise in self-employment (up 543,000 from last March). Self-employed workers count on the household survey -- but obviously not as employees. In fact, the rise in self-employment is probably understated. I suspect many contract workers count themselves as employees since they work entirely or primarily for one company. But if they receive a 1099 at the end of the year instead of a W-2, it makes them legally self-employed.

Companies have moved toward using contract workers in part because it provides more flexibility but also because it means they don't have to pay benefits. The flexibility provides an obvious advantage in a fast-moving economy. Avoiding benefit costs looks more attractive, too, since benefits are rising nearly three times as fast as wages and represent a growing share of employment costs. The 6.9% rise in benefit costs last year marks the largest since 1984. In fact, benefits make up more than 29% of total employment costs.

The drop in the unemployment rate spreads out demographically. The largest decrease was for black adult males (to 9.2% from 10.9%), but the rates declined for almost every category except black teenagers, where the sample size is too small to take monthly movements seriously.


  On a disheartening note, the average duration of unemployment rose to 19.5 weeks. Although the median duration held at 9.3 weeks, 21.5% of the unemployed have now been out of work for more than 26 weeks. Discouraged workers (who are counted not as unemployed but as not looking for work because they doubt they can find jobs) held at 0.3% of the workforce, down slightly from a year ago.

The employment report continues to underscore the importance of education. Individuals with a bachelor's degree or higher had an unemployment rate of only 2.4%, while high-school dropouts registered a rate of 7.8%. Even an uncompleted higher education helps. Those with some college experience had a 4% unemployment rate, vs. a rate of 4.8% for high-school grads with no college.

The core consumer price deflator, Federal Reserve Chairman Alan Greenspan's preferred indicator of inflation, rose 0.2% in February and has increased 1.6% from a year ago. The total deflator (including food and energy) is up 0.3% on the month and 2.3% over the last 12 months. The 12-month change in the core deflator has come in at 1.5% to 1.7% for the last seven months. The weaker payroll data and mild inflation suggest the Fed will refrain from accelerating its tightening at the next meeting. The central bank needs a couple of months of strong employment gains or rising inflation to justify faster rate hikes. We still expect the Fed funds rate to close the year near 4%.

Wyss is chief economist for Standard & Poor's

Before it's here, it's on the Bloomberg Terminal.