Operations

At the end of 1986, Nucor's steel operations, encompassing 16 steelmaking and fabrication plants at 10 locations around the United States, accounted for 99% of the company's sales (see Exhibit 9). By far its largest operating division was Nucor Steel, which that year produced 1.7 million tons of steel bars, angles, light structural shapes and alloys at steel mills in South Carolin. -.ebraska, Texas and Utah. Nucor was ranked as the second most productive steelmaker in the woric in 1985 based on its annual tonnage per employee (981 tons), behind Tokyo Steel but ahead of the largest U.S. minimill, North Star (936 tons), and the largest integrated steelmaker U.S. Steel (479 tons).10 Nucor benchmarked its plants against its leading competitors, particularly Chaparral, as well as other Nucor plants in the same division. It had a policy of letting competitors visit its operations as long as they reciprocated.

Nucor Steel sold two-thirds of its output to external customers and one-third internally. Vulcraft, whose joist plants originally impelled Nucor to integrate backward into steelmaking, accounted for three-quarters of internal sales. In 1986, Vulcraft operated six plants, four of which were located very close to Nucor's four steel mills, and had sold 450, 000 tons of steel joists—about 30% of the U.S. market—and 175, 000 tons of steel deck. Vulcraft sourced 95% of the steel bars that it welded into joists from Nucor Steel, at a discount of $10 per ton relative to market prices that was justified on the basis of high volumes and the absence of credit and collections problems. Vulcraft purchased the flat sheet that it grooved and corrugated into deck from outside suppliers. The remainder of Nucor's internal steel sales were channeled to smaller downstream operations that made cold-finished bars, grinding balls, bolts, bearings and machined steel parts.

⁹F. Kenneth Iverson, " Effective Leadership: The Key is Simplicity, " in Y. K. Shetty and V. M. Buehler, eds., The Quest for Competitiveness (New York: Quorum Books, 1991): 287.

¹⁰ I ron Age (May 2, 1986): 58B1.

Nucor's dependence on external sales of steel had increased dramatically since the early 1970s, when they accounted for only 10% - 20% of total production. Although Nucor's four steel mills were geographically dispersed and each tried to maintain an inventory of 25, 000 tons, they occasionally shared orders when they could not fulfill them alone. Service centers and distributors constituted their primary customers. Prices were set centrally on an FOB. (freight on board) basis, unlike the delivered cost prices quoted by-integrated steelmakers and some minimills. Nucor did not allow discounts for preferred customers or, since 1984, on large outside orders. The elimination of quantity discounts reflected the computerization of Nucor's order-entry and billing systems, which reduced fixed order-processing costs, and Nucor's intention of differentiating itself, especially vis-a-vis imports, by letting its buyers maintain lower inventories and order more frequently. Nucor remained willing, however, to offer temporary discounts of up to $20 per ton when it started up a new steel mill.

Nucor tried to avoid haggling about prices on the input as well as the output sides by coordinating the purchase of steel scrap through an independent purchasing agent, David Joseph, Inc. Each mill was assigned a David Joseph representative who, upon receipt of an order for scrap, first checked other accounts to see if economies could be realized by pooling orders and then looked for sources within the region. Nucor paid David Joseph a fixed commission per ton of scrap.

Nucor had long focused its operations on production rather than procurement or marketing. This focus had forced it over the years to think very hard about how it recruited, trained and motivated production workers. The results were widely acclaimed. In the words of investor Warren Buffett, " It is the classic example of an incentive program that works. If I were a blue-collar worker, I would like to work for Nucor." ¹¹ Nucor attracted a large number of applicants for each job that it advertised. Iverson's favorite story involved eight new jobs that had opened up at the Darlington steel mill in 1985. By 8: 30 a.m., the 1, 300 applicants had created such a traffic jam that the state police had to be called in for assistance. Unfortunately, the police were short-handed: three of their officers were already at Nucor applying for jobs.

Nucor selected and trained all employees below the level of department head (accounting for about 95% of its total compensation budget) on a decentralized basis. Each plant general manager administered a psychological test to prospective employees that sought to identify goal-oriented, self-reliant people. Previous steelmaking experience was not, for most jobs, an important selection criterion. Once selected, production workers were trained by foremen to perform multiple functions. After a two-month training period, each employee was assigned to a 20- to 40-person production group that performed a discrete task, such as melting scrap into steel, rolling steel; or finishing steel by straightening it. On-going training was provided by the more experienced members of the production group. Most promotions on the shopfloor were made from within based on performance and peer evaluations. Production shifts were eight hours long, with an average workweek of 42 hours in a rotating pattern. Plants ran continuously for six days a week, with the seventh day reserved for maintenance.

¹¹ Fortune (December 19, 1988): 58.

Nucor's compensation systems were designed to reward production groups, rather than individual workers, for exceeding predetermined productivity and quality standards. The standards were typically based on experience rather than formal time studies and were not revised unless there was a major machinery change. Less than 5% -10% of the standards were changed in any given year. No bonus was paid when equipment was idle. Since incentive bonuses could average 80% -150% of base wages, cash compensation was slightly higher on average for Nucor's nonunionized production workers than for unionized steelworkers at integrated steelmakers, even though Nucor's base wage per hour was significantly lower. Nucor reinforced these rewards with stiff penalties: anyone late for a shift lost a day's bonus, anyone who missed a shift lost the bonus for the week, and if a group fell short of its productivity goal, all its members lost their bonuses for the week. It further reinforced the relationship between pay and performance by paying bonuses every week (with green checks) and by continually reminding workers of their progress. At the entrance to each plant hung a giant board that depicted each group's weekly productivity gain, their bonuses, the plant's performance relative to the target of 25% return on assets, the company's return on equity for the month, and the latest stock price.

Employee turnover at Nucor was about 1% - 5% per year, compared to an average of perhaps 5% to 10% for the U.S. steel industry as a whole. Turnover was highest among new production workers to Nucor but declined dramatically, with concomitant increases in productivity, after some quick departures. Nucor claimed that it never fired workers who performed their jobs up to reasonable expectations. Nor did it lay them off when demand dropped; instead, it shortened its work week as necessary. During a short week, production bonuses remained in place but might be based on one or two fewer days of work, reducing the average worker's total pay by 15% - 20%. Under the company's " Share the Pain" program, department heads' pay might fall by 30% - 40% at such times, and officers' by 60% - 70%. Between 1981 and 1982, for example, Iverson's compensation fell from $276, 000 to $107, 000, landing him near the bottom of the Fortune 500 on that measure. By comparison, the average compensation for CEOs of the seven largest integrated steelmakers dropped from $708, 000 to $489, 000 during the same period.

Although Nucor's operations were decentralized down to the plant level, there was considerable interplant communication. Some of this communication occurred through formal channels, such as the three meetings with corporate executives that all plant general managers attended each year and quarterly function-oriented meetings. Most communication, however, took place through informal channels, such as the one- to two-day visits that managers and workers frequently paid to other plants. Informal communications was encouraged by broad dissemination of data on the performance of individual operating units and by the incentive system, which motivated operating units to beat their performance targets and to take an interest in overall corporate performance as well. Some of the mechanisms used to ensure that information was transferred from old plants to new ones are described in the next subsection.