The employment report is supposed to be the Labor Dept.'s monthly snapshot of the U.S. labor market. The May installment, though, looked like a double exposure: Payrolls rose less than expected, a sign that economic growth is moderating to a less inflationary pace. But the jobless rate dropped sharply to a level that suggests wage and price pressures are just around the bend. Which image is correct?
Stick with slower, noninflationary growth. Besides, the government's payroll survey has always been more dependable than its survey of households, from which the data on the labor force and the unemployment rate are derived. That's especially true now, since Labor is still working out the kinks resulting from its switchover a few months ago to a new method for measuring joblessness that tries to remove gender bias from the household survey.
The payroll data show that businesses added 191,000 workers in May, well below the hefty increases of 358,000 in April and 379,000 in March. What's more, that gain includes the return of the 70,000 Teamsters who were on strike in April. The May payroll additions were the narrowest in nine months, with only 50.6% of 356 industries putting on new workers, compared with 62% in the two previous months.
The smaller May job gains, taken together with weaker April readings for the index of leading indicators, factory orders, and consumer spending, support the notion that growth in the economy is cooling down to a pace that will be to the liking of both the financial markets and the Federal Reserve.
Without a doubt, labor markets are tighter now than a year ago. In fact, spot shortages of labor are beginning to crop up in a few categories. Moreover, the weak California economy is single-handedly adding 0.3 percentage points to the national jobless rate, suggesting that the overall rate is understating labor-market tightness across the rest of the country (chart).
The plunge in the unemployment rate, however, from 6.4% in April to 6% in May, just didn't happen. Labor Dept. officials admit that the reliability of the new household data is questionable. The main problem is a lack of sufficient data history, without which proper seasonal adjustment cannot be made. Indeed, under the old measure, May joblessness fell to 6.1%, a surprising result given that the new rate has typically run half a percentage point higher than the old one.
Moreover, the new data show that the labor force has risen only 0.8% in the past year. That runs counter to the growth in the working-age population, which has increased twice as fast. In particular, the numbers say the labor force hasn't grown at all since February. No way. If the work force had risen at a more realistic pace, the May jobless rate would have been much higher.
As Bureau of Labor Statistics Commissioner Katherine Abraham noted after the data for May were released, large movements in the unemployment rate in a single month often are partially reversed in the following month. That should be the case this time as well, suggesting far less upward pressure on wages than the May jobless rate would imply.
To date, wage growth shows little sign of accelerating. Average hourly earnings rose 0.5% in May, to $11.11, but so far in the second quarter, hourly pay has increased just 2.6% from a year ago. That annual pace is up just slightly from 2.4% in the second quarter of 1993 (chart).
What's more, all of that pickup in the past year occurred in manufacturing, where productivity is growing at a near-record pace and where unit labor costs are falling. Pay growth in the larger service sector has gone nowhere for two years.
But even as wage growth stays low, workers generally are generating plenty of income. That's because they are working longer hours. The workweek rose to 34.9 hours in May, a seven-year high. As a result, average weekly earnings during the past year are up 3.3%, faster than inflation.
Moreover, the quality of jobs is improving. Businesses, for instance, are relying less on temporary help--employees who don't accrue costly benefits or seniority. In the past year, only 336,000 new jobs arose in what the Labor Dept. labels "personnel-supply services," or temporary help agencies. This sector chalked up 13.1% of all new jobs created. While that percentage may appear to be impressive, such temp jobs accounted for about 50% of all job growth two years ago (chart).
And temps are no longer the stereotypical receptionist working the front desk. Economists at the Federal Reserve Bank of Chicago looked at the nature of temp jobs. In particular, they examined temporary jobs classified as industrial workers--employees who handle heavy machinery, run production lines, and move goods. While some industrial workers may work at wholesale or transportation companies, most probably are employed in manufacturing.
The Chicago Fed researchers estimate that at the beginning of this expansion, about 100,000 new factory workers a year were hired through temp agencies. That rate has slowed to 85,000 in the past year, as factories began to add to their payrolls outright.
What the analysis suggests is that instead of paring 62,000 production jobs since 1991, the factories have actually added about 200,000 jobs. And that doesn't include clerical and white-collar temps who work at manufacturing companies.
If the May employment report was a bit blurry, the latest data on factory activity, consumer spending, and installment credit offer a more focused picture of the outlook. What's in the viewfinder? Evidence of slower growth as the economy begins feeling the effects of higher interest rates.
The loss of 2,000 factory jobs in May and the decline in the factory workweek from 42.2 hours to 42.1, for instance, suggest that industrial production was weak last month. Slowing demand explains why manufacturers have little need to speed up assembly lines.
Factory orders slipped 0.1% in April, and since January, they are up just 1%, compared with a 5.6% gain in the preceding three months. Inventories, however, edged up only 0.2% in April. So factories aren't stuck with a stock overhang that would foreshadow steep output cuts.
Tight inventories can backfire sometimes, though. The recent drop-off in sales of motor vehicles is partly the result of a shortage of popular models as well as the impact of higher interest rates on car loans and leases. Sales of U.S.-made cars and light trucks fell to an annual rate of 12.6 million in May, down from 13.1 million in April (chart).
Sales at department and chain stores slacked off in May, as well. The Johnson Redbook Report says that sales last month fell 2.6% before rebounding 3.1% in the first week of June from the May average.
The shopping slowdown means that gains in installment credit should taper off. In April, however, consumers added a gargantuan $8.9 billion in debt--the most in more than nine years, including a $5 billion surge in revolving debt.
Why are consumers feeling more comfortable about pulling out their credit cards? A stronger sense of job security probably helps, as does better income growth. Debt as a share of disposable income stood at 16.7% in April. While that ratio is on the rise, it is still quite manageable.
The expected gains in jobs and incomes mean that debt loads should remain benign. So, too, the expansion should stay on its course of slower growth. That trend isn't very glamorous, but it is the surest way to keep inflation out of the economic picture.