SMART Modular Technologies, Inc.

views updated

SMART Modular Technologies, Inc.

4211 Starboard Drive
Fremont, California 94538
U.S.A.
Telephone: (510) 623-1231
Fax: (510) 623-1434
Web site: http://www.smartm.com

Public Company
Incorporated: 1988
Employees: 1,079
Sales: $707.4 million (2006)
Stock Exchanges: NASDAQ
Ticker Symbol: SMOD
NAIC: 334119 Other Computer Peripheral Equipment Manufacturing

SMART Modular Technologies, Inc., is a Fremont, California-based designer, manufacturer, and supplier of about 500 computer component subsystems, including memory modules and cards, flash modules and cards, controller boards for liquid-crystal displays, and embedded computing subsystems. SMART sells to many leading original equipment manufacturers (OEMs) in the computer, networking, and telecommunications fields, including Cisco Systems, Dell, Hewlett-Packard, IBM, Motorola, and Xerox. The company operates 11 plants around the world.

FOUNDER IMMIGRATES TO AMERICA: 1980

SMART was founded by Ajay Shah, born in Uganda but of Indian descent. Although his family was well established in Uganda—Shah was a member of the third generation to call the country its home—they were driven out when dictator Idi Amin took power and instituted his own brand of racist polices, which included the removal of all people of Asian extraction. In 1971 the family returned to India, where Shah received a boarding school education and a degree in mechanical engineering from the University of Baroda. He then went to work for Larsen and Toubro, two of the top engineering firms in India, but he never felt comfortable with India’s stifling caste system and pervasive bureaucracy. In 1980 he immigrated to the United States to attend Stanford University and two years later received a master’s degree in engineering management.

After Stanford, Shah took a job in 1982 at Advanced Micro Devices Inc. (AMD), maker of computer processors, where he spent six years and impressed management with his ability, dedication, and winning personality. At AMD he gained a thorough understanding of the business, from wafer design and fabrication to product marketing, lessons that he would later apply to SMART. Shah’s next position was product manager for a new U.S. subsidiary of Japan’s Samsung Electronics Co. Ltd.; his role was to manage Samsung’s line of computer memory products as well as to develop business plans for new semiconductor memory product lines. In 1988, when there was a shortage of DRAM (dynamic random access memory), for use in personal computers (PCs) and workstations, Shah proposed that Samsung begin manufacturing memory modules, which would give personal computer manufacturers an opportunity to offer PCs at different price points, depending on how much memory the units offered. Moreover, only a handful of companies were involved in the manufacture of memory, providing a clear opening, but Samsung did not consider the market large enough to warrant the investment. Shah disagreed and in 1988 he resigned to start his own company to make memory modules.

Cofounding SMART was Shah’s wife, Lata Krishnan, and an AMD colleague, Mukesh Patel, who headed engineering. Although Krishnan was born in India, her family had also fled East Africa. The couple met in London when she was attending the London School of Economics. She took a position with Arthur Anderson & Co. and then became controller at Montgomery Securities Inc. She quit her job to become chief financial officer at SMART, and drew up the company’s business plan, which was essentially a downscaled version of the Samsung plan but tailored for a small start-up company. She also raised money, primarily from East Africans her father knew. Altogether, the founders cobbled together $110,000 to launch the San Francisco Bay–area company. They did not seek out venture capital money, thus maintaining their freedom and freeing themselves of the pressure to take the business public too quickly.

SMART commenced operations in 1989, employing about 30 people in a 28,000-square-foot plant in Fremont. The company looked to OEMs for its customers, uninterested in direct sales to consumers. SMART’s timing proved fortuitous as Apple Computer had begun designing memory modules into its systems and needed a supplier. Apple became the first major OEM customer. Ironically, another early customer was Samsung. Other OEMs and semiconductor manufacturers soon followed, so that after just two months the start-up became profitable, helped to a large degree by Shah’s frugalness. Most of the furniture in the corporate headquarters was bought used, and the company even shipped its products in the same plastic trays that carried the memory modules through the assembly line. “We’re kind of cheap,” Shah told Forbes in a 1996 interview, “but tell me what is wrong with this?”

SALES SURGE: 1993

When he was not pinching pennies, Shah was thinking ahead, quickly realizing that DRAM would become price sensitive, and so the company began looking to the “fringes” to find more growth and gain some diversity. He told Electronic Buyers’ News in a 1997 interview, “We knew we didn’t want to be involved in the middle of a commodity market, so we worked hard to deliver more technical differentiation and more value-add.” In 1991 the company expanded its product offerings to include specialty memory modules and PC card memory products. Sales approached $16 million in fiscal 1992, when the company turned a profit of $292,000, but business began to explode. In 1993 SMART introduced a suite of PC card communications products, the addition of which helped to increase revenues to $80.8 million and net earnings to $3.3 million.

To meet demand, SMART opened a second plant in Puerto Rico and sales offices around the world, as sales continued to grow at a rapid pace. The company recorded sales of $163.8 million in fiscal 1994 and net income $6.2 million. SMART gained more diversity in July 1995 when it acquired Apex Data, Inc., maker of wireless and wireline PC card data/fax modems and connectivity products, some of which included flash memory. Nevertheless, most of SMART’s accelerated growth at this time was due to the tremendous demand for DRAM memory, caused mainly by the new Windows 95 operating system introduced by Microsoft, which required an increasing amount of memory. In turn, the increasing demand for PCs for business and home use meant a greater need for computer memory. Hence, SMART’s revenues grew to $274.6 million and net income more than doubled to $12.6 million in fiscal 1995.

COMPANY PERSPECTIVES

Our success is derived from a customer-focused approach characterized by a commitment to quality, advanced technical expertise, fast time-to-market, build-to-order flexibility and high quality customer service.

Shortly after fiscal 1995 came to a close at the end of October, SMART conducted an initial public offering (IPO) of stock. With Donaldson, Lufkin & Jenrette Securities Corp., Cowen & Co., and Oppenheimer & Co. acting as comanagers of the underwriting group, SMART raised $33.5 million. The IPO’s timing was poor, but the offering held as the price of memory chips plummeted. Investors mistakenly believed that SMART’s prospects were contingent upon high chip prices, not realizing that the key to the company’s success was high volume coupled with service. SMART was also on the verge of adding PC assemblers Gateway 2000 and Dell Computer Corp. to its roster of clients.

Although SMART’s stock would early on trade around the $8.50 mark on the NASDAQ, it quickly gained in value as the company continued to post strong results. In fiscal 1996 revenues topped $400 million and earnings totaled $25.1 million. To keep up with demand and support growth in Europe, SMART opened a new 25,000-square-foot manufacturing facility in East Kil-bride, Scotland, its first plant located outside the United States and Puerto Rico. The company also continued to diversify. In July 1996 it became involved in the embedded computer market by acquiring RISQ Modular Systems, Inc., a California designer and manufacturer of embedded processor modules, which performed just one or a small number of dedicated tasks. The market for such products was growing rapidly because they were critical components in a wide range of devices, including laser printers, scanners, cable and satellite TV set-top boxes, video games, medical imaging analyzers, computer guided missiles, and paint mixing systems. Because OEMs were under increasing pressure to get new products to the market, they were turning to outside companies such as RISQ to design and manufacture embedded systems for them.

In October 1997 SMART reorganized its business, establishing two units to house its new product lines: the embedded-computer division and the communications-product division. The company’s decision to diversify proved wise. Even as the DRAM market fell off in the late 1990s, SMART was able to enjoy continued growth, albeit not at the pace it had achieved through most of the 1990s. In fiscal 1997 the company generated sales of $694.7 million and net income of $45.4 million. The following year, when DRAM prices tumbled, SMART was still able to increase revenues to $714.6 million and earnings to $51.5 million, but the price of its stock fell to $14, a far cry from the $44 it fetched at its peak. To achieve some cost competitiveness SMART opened its first Asian plant in fiscal 1998, located in Penang, Malaysia.

SOLECTRON ACQUIRES SMART: 1999

As DRAM prices continued to languish in fiscal 1999, SMART sought ways to make up the loss of revenue from its core memory module business. In August 1999 it acquired Compaq Computer Corp.’s embedded and real-time product line. It also began looking to become involved in contract manufacturing, possibly setting the stage to becoming a “box builder.” SMART was then approached by Solectron Corp., the world’s largest contract manufacturer, about a possible acquisition. An agreement was reached and in September 1999 SMART was acquired by Solectron for $2 billion in stock. At the time, Shah and his wife owned a 22.4 percent stake in SMART. It was a deal that made sense for both companies. SMART needed the security of a larger company, and Solectron valued SMART’s complementary customer base and engineering and design capabilities. While Shah turned over the running of SMART to the company’s president, he took over the management of Solectron’s Technology Solutions unit, essentially a unit that worked with clients to develop product concepts. It then designed, built, and serviced the products on an outsourced basis.

The addition of SMART was just part of an acquisition spree for Solectron that continued into the early 2000s, as it used its high-flying stock to finance the deals. A downturn in the high-tech sector, however, adversely affected business. Despite increasing sales to $18.7 billion in fiscal 2001, Solectron posted a loss of $124 million. The company cut jobs and closed down facilities. SMART’s East Kilbride plant, for example, lost 150 jobs in 2002. All told, thousands of employees would be put out of work and the company was forced to take write-offs that amounted to billions of dollars.

KEY DATES

1988:: SMART Modular Technologies is incorporated to manufacture memory modules.
1989:: Operations begin.
1995:: Company is taken public.
1999:: Solectron acquires SMART.
2004:: Founder and partners reacquire the company.
2006:: SMART is taken public again.

Part of Solectron’s retrenchment called for a return to focusing on its core business. Hence, SMART was expendable, and in February 2004 Shah, who had left Solectron the previous year, stepped forward with an offer to buy back the company he had cofounded. Having cashed in more than $63 million in Solectron stock, while retaining 5.6 million shares, he and his wife had formed a private equity company, Shah Capital Partners. He teamed up with a pair of similar firms, Francisco Partners and Texas Pacific Group, to make a successful $100 million bid for SMART, just 5 percent of the price Solectron had paid five years earlier. Francisco Partners was a technology company investment vehicle for Sanford Robertson, cofounder of Robertson Stephens investment bank, while Fort Worth, Texas-based Texas Pacific was run by David Bonderman, part of the inner circle of the billionaire Bass family.

Shah resumed his role as SMART’s chairman. Serving as chief executive was Iain MacKenzie, a seasoned executive with 25 years of experience who was well familiar with the SMART operation. He had established and run the company’s European subsidiary and had also served as vice president of SMART’s worldwide operations before becoming the company’s president in 2002. He and Shah now worked together to rebuild the business. SMART lost $29.8 million on sales of $892.8 million in fiscal 2004, but returned to profitability the following year with earnings of $26.2 million, despite a decrease in revenues to $607.3 million.

With the balance sheet in order, SMART was once again ready to tap the equity market. In February 2006, the company was taken public a second time, raising $76.1 million for corporate use while its investors took some profit. For his part Shah sold 1.6 million shares, resulting in a $13.4 million payday. The company continued its strong recovery, improving sales to $707.4 million in fiscal 2006 and enjoying a 23.3 percent increase in earnings to $32.2 million. Business continued to thrive in the first quarter of fiscal 2007, and the company took steps to make a secondary offering of stock, the proceeds to benefit selling shareholders, such as Shah Capital which indicated it planned to sell 2.2 million of the 6.9 million shares it owned.

Ed Dinger