Soon after it was announced that payroll employment expanded by only 21,000 in February, the bond prices rallied sharply (yields dropped) and the stock market pulled back. Investors have become concerned that the weaker than expected employment picture (coupled with high gasoline prices) does not bode well for the consumer economy. For the most part, the post recession employment pattern is tracking as we envisioned some twelve months ago. (See ‘The Good, The Bad, and The Ugly’ of January, 2003.) In that piece we demonstrated the strong correlation between non-residential expenditures (capital spending on structures, equipment, and software) and employment. We have updated our data to compare the annual changes in capital spending and hours-worked in non-farm businesses. (See Chart 1).

Historically, hours-worked have improved at a greater rate than capital spending during recoveries. On a rate of change basis, both troughed in 2002-QI. (On an absolute basis, capital spending bottomed in 2003-QI, and hours-worked troughed the following quarter.) Since then, the year-over-year change in hours-worked had been accelerating upwards as expected, but recently lost the momentum we had expected based on the increases in non-residential capital spending.

In part this de-linking between capital spending and employment may be due to the technology surge we observed in the late 1990’s, e.g., better productivity. In the fourth quarter, non-farm business sector productivity improved by 5.4% over the previous year resulting in a decline in labor unit costs of 1.7%. On the other hand, non-durable manufacturing productivity has lagged.

PRODUCTIVITY, COSTS & HOURS WORKED

As one can see from the above table, non-durable manufacturing has not been achieving comparable productivity increases in recent years and in 2003 experienced an increase in unit labor costs of 3.9% vs. a 2.1% decline for durable manufacturing. Much of this sad story is related to declining output in this sector leading managements to reduce hours worked by 2% in the fourth quarter (from the third quarter) and 4% for all of 2003!

Between last September and February non-durable goods manufacturing has lost 75,000 jobs. (This compares to a gain of 8,000 for durable goods manufacturing.) Almost half of those jobs were within the food, beverage, and tobacco industries. The other big losers were textile mills (13,800), and chemicals & plastics (10,600). Clearly, the production levels for food, beverage & tobacco declined significantly in last year’s fourth quarter and continue to be weak. Production rates for many of the “import impacted” products such as apparel, textiles and printing are still suffering, albeit at a lesser rate. In other words, there is some encouragement that these sectors will not be such a drag on employment.

It was the food, beverage, & tobacco arena that experienced the largest sequential production decline in the fourth quarter and was the major source of job losses as well. Since last September, this sector has lost 33,300 jobs, with food processing accounting for 29,400. While some of the declines may be associated with inclement weather, most are associated with bovine spongiform encephalopathy (BSE), or “mad cow disease”. The cessation of imports of Canadian cattle, due to the discovery of BSE last May, and a depletion of cattle-on-feed in the second and third quarters of last year created a lack supply in the fourth quarter, and workers were laid-off. Domestic meatpacking houses (the largest employer within the food group) reported a 12% decline in beef production and a 6% shortfall for all red meat over the four months ending January 2004. Beef production is expected to decline through mid-year and then experience a recovery. This industry saw an approximate 11,400 reduction in employment (not seasonally adjusted) from September through January. (February data is not available). Also of significance was a 6,400-worker decline at bakeries; employment has been dropping steadily since 2000 thanks to Dr. Atkins.

The next few quarters will probably see continued declines in these sectors, but there is hope that meatpacking, chemical & plastics, and printing industries will return to the plus column. And if China does revalue the yuan, our textile industry might stage a recovery of sorts. Still, the bottom line is that the employment statistics are going to improve, but at a much slower rate than our model would suggest. After June, however, there is the potential for employers to “put on the steam” as inventory levels are replenished and consumers spend their tax windfall.

While investors are now fretting that the consumer economy may encounter some softness during the second half, the profit picture for the remainder of this year appears quite healthy. The ratio of unit selling prices to unit labor costs, which widened during 2003, will expand over the next three quarters whereupon it should begin to plateau. (See Chart 2). This ratio is an excellent proxy for non-financial corporate pretax profit margins. We expect profit margins to expand into the summer, and the consumer economy to remain buoyant caused in part by the recent decline in mortgage rates. This could spur another re-financing boom if ten-year Treasury bonds remain at current yields.

John K. Dolan
Brian E. Schaefer

S&P 500: 1124
March 11, 2004