For more than 20 years, we've been counting on transportation deregulation, just-in-time strategies, supply chain management practices and collaboration to increase inventory turns and reduce the amount of inventory we keep on hand. But it turns out that those strategies may not have worked as well as we thought.

A study by James Ginter and Bernard LaLonde of The Ohio State University that focused on companies' inventory-management performance between 1979 and 1999 found that although some industries have made substantial improvements in their inventory levels as a portion of their costs, others have shown no improvement, and in some cases, inventory levels have actually grown. Given that businesses have, at least in theory, spent enormous amounts of attention on improving their supply chains in recent years, LaLonde says, those findings are counterintuitive.

Ginter and LaLonde looked at the 10-K financial reports of publicly traded companies compiled by Standard & Poor's. They eliminated service companies and those with less than $100 million in sales in 1999, then collected data on the ratio of inventory to the Cost of Goods Sold (COGS) from the balance sheets of the remaining companies. (The researchers chose COGS rather than sales to eliminate the effects of gross margins on the data.) They then analyzed trends in industry sectors that included at least 30 companies.

What they found was that for the 14 industries where they could analyze total inventory trends, overall inventory ratios decreased in nine. The adjusted total inventory levels for the apparel and food products industries grew. For the remaining three—furniture, medical products and other consumer packaged goods—inventory showed no significant change.

The results also showed that inventory levels for raw materials and work-in-process (WIP) declined in the majority of industries, but finished-goods inventories increased in seven of the 14.

"Our original research was driven by a hypothesis that most people looked at inventory as composed of those three pieces [raw materials, work in process, and finished goods]," LaLonde says. "Logic told us that those three could behave differently."

The results of the survey bear that out, and the researchers have a theory for this, too. "Our hunch is that companies have greater control over their internal processes and [therefore] can control raw materials and WIP," says Ginter. "Finished-goods inventories are driven in part by the customers."

Interestingly, the researchers discovered that although the retail/wholesale industry has decreased its total amount of inventory, businesses that sell to retailers have increased their inventory levels. In addition, the proliferation of stock-keeping units (SKUs) among consumer goods manufacturers may have offset some of their efforts to improve inventory velocity.

The paper, "An Historical Analysis of Inventory Levels: An Exploratory Study," can be found at http://fisher.osu.edu/scmrg.