Why New Production Plants Are Vulnerable to Closings

March 3, 2003

Why New Production Plants Are Vulnerable to Closings

By CLARE ANSBERRY Staff Reporter of THE WALL STREET JOURNAL

PITTSBURGH -- True or false? A big, older plant that pays above-average wages and is capital-intensive is more vulnerable to a shutdown than a smaller, newer one that pays lower wages and is less capital-intensive.

Answer: False.

Plant closings occur all the time. Studies show on average over a five-year period, more than 32% of U.S. manufacturing plants, or 6% a year, are shuttered. When the economy is weak, there are more closings as companies try to slash costs. A logical conclusion would be older plants with higher wages would be high on the hit list. There have been hundreds of tire, steel, machine-tool, auto, textile and furniture plants -- all older with high-paid workers -- closed in recent years.

But economists J. Bradford Jensen, with the U.S. Census Bureau, and Andrew B. Bernard, at the Tuck School of Business at Dartmouth College in Hanover, N.H., found that isn't typically the case. Looking at about 117,000 plant closings, from 1977 to 1997, which includes several economic cycles, they found, among other things, that shutdowns occur less frequently at larger, older, more capital-intensive plants. The plants studied included various industries and regions of the country, as well as union and nonunion shops.

Longevity reflects certain strengths that offset higher costs. A 30-year-old plant lasts three decades because somebody inside was doing many things well: making the right product with an efficient process and keeping machines running well. Typically, they have invested heavily in capital equipment and technology, which enhances productivity. Moreover, their work force is skilled, which means experienced, and that usually means high-paid. "Plants that can't compete go out of business. The ones that can compete survive and grow older," Mr. Jensen says. Moreover, he expects those findings to hold true for the foreseeable future. "Big, old capital-intensive, skill-intensive plants are more likely to survive going forward."

Even in a strong economy, newer plants are vulnerable. Within five years of start-up, half of new factories disappear. Ten years later, only one-third are left, notes the Manufactures Alliance, a trade group in Arlington, Va. But in a weak economy, the performance pressures are even greater on new operations. For some companies trying to streamline, it makes more sense to close a newer plant and move production to one that already has an established supplier and distribution network.

Newer plants may employ more experimental processes that don't pan out, or target a market that doesn't materialize, both of which seem extravagant in a weak economy. ConocoPhillips Inc. has announced it was closing its two-year-old "revolutionary" $125 million carbon-fiber operation in Ponca City., Okla., which was supposed to make eight million pounds of a material touted as stronger and lighter than plastics and metals. A combination of uncertainties in that market and operating and technology difficulties killed the project.

Another issue in determining longevity is exports. Nonexporting plants are 15% more likely to close than plants that export, the census study says. Exporting, then, would seem to create a stronger operation. Instead, it is the other way around. Only a strong operation can export. Getting into the export market isn't like flipping on a switch. There is a lot of work establishing distribution channels overseas, and dealing with legal, administrative, political and regulatory issues. A struggling operation usually doesn't have the resources to jump into, or at least succeed, in the export market. "Good plants become exporters. It's not that exporting makes plants better," Mr. Jensen says.

Size plays a role, too, in determining which plant survives, and which doesn't, especially when high volume is critical to keeping costs down. Ball Corp. announced last week it was closing its Bytheville, Ark., welded food-can-making plant, the company's smallest both in terms of employment and production, and shifting output to a bigger plant in Arkansas. "We have to have very large plants to spread the cost of capital over our operations," Ball spokesman Scott McCarty says. "Even if a smaller plant has the same wages and equipment, it can only put out so much volume."

Ball is also installing food-can production lines in plants that make only beverage cans and is considering making plastic bottles in some metal-only plants. At multiproduct plants, production can switch to a gangbuster product to offset another product whose market has hit the skids.

One variable wasn't measured in the study simply because it is hard to quantify: management. Good management hires the right workers, pays them competitive wages, buys the best equipment and chooses the right products and end markets. "It's a key unmeasured input," Mr. Jensen says.

Write to Clare Ansberry at 1