Financial performance of computer network and information technology services companies in bull and bear markets.

Macy, Anne ; Terry, Neil ; Walker, Jean 等

INTRODUCTION

The information technology sector has transformed the economy and changed the basis of competition (Sampler, 1998). Information technology boosts the efficiency of the decision-making process and is perceived by many executives as an integral part of their business strategy (Molloy and Schwenk, 1995; Bartholomew, 1998). Investors have struggled to comprehend the potential and the limitations of information technology companies as the industry has continued to evolve over time. Not surprisingly, the volatility of stock prices for information technology firms has been extreme as many companies struggle to survive in the next few years after reaching a peak stock valuation. On March 10, 2000 the NASDAQ composite peaked at an intra-day high of 5,132 and declined to half of its value within a year before finding a bear market bottom on October 10, 2002 with an intra-day low of 1,108. The excessive rise and fall of information technology companies offers a unique opportunity to evaluate industry nuances associated with bear and bull markets.

The purpose of this research is to compare the stock market performance of multiple computer network and information technology services companies across six information technology eras. The six period classifications are the browser era, Y2K era, post-Y2K era, post9/11 era, outsourcing era, and mobile/wireless era. Cisco Systems (CSCO), 3Com (COMS), Ericsson (ERIC), Nortel Networks Corporation (NT), and Yahoo Inc. (YHOO) are the five computer network and information technology services firms included in the study. The organization of this manuscript divided into five sections. The first section offers a discussion on the literature related to the financial performance of information technology companies. The next section offers background information relating to the six information technology eras applied to this study. The third section discusses the computer network and information technology services industry and the five specific companies that are the focus of this study. The fourth section presents data and methodology. The fifth section puts forth results from the application of a nonparametric technique to compare stock market returns across different information technology eras for the six companies. The final section offers concluding comments.

REVIEW OF THE LITERATURE

Academic research identifying structural economic changes that influence stock prices mostly focuses on major crashes in the history of financial markets (Higgins & Osler, 1997; Allen & Gale, 2000; Cocca, 2005). Although a relatively new topic for the information technology sector, there are numerous studies in finance theory that focus on the development of speculative bubbles and stock market volatility (Camerer, 1989; Allen & Gale, 2000). Stock market volatility is explained by various approaches, which differ in essence according to assumptions made with regard to market efficiency (Sornette & Malevergne, 2001). Stock market performance of information technology companies reveals the sector has greater volatility than most other economic sectors (Demers & Lev, 2001; Ofek & Richardson, 2003; Kamssu, Reithel, & Ziegelmayer, 2003). Terry, Macy, and Abdullat (2010) find a correlation of stock prices for vertically integrated technology companies in a down market but bull markets are not highly correlated within the vertically integrated firms.

Cocca (2005) puts forth one of the few studies exploring potential reasons for the stock market volatility of information technology companies. The study uses a broad media database to analyze the informational and media environment surrounding the market highs for technology stocks and explores potential trigger events that could cause an Internet bubble to burst. Two key informational event triggers are public awareness of the human genome research results and the publication of a study by Barron's magazine about Internet companies' burn rates. Cocca (2005) concludes diffusion data of the informational events show a long-term impact of the Barron's study on media, financial analyst and consequently investor focus.

Researchers are becoming more and more interested in studies relating IT investment and firm performance (Im, Dow, & Grover, 2001). The studies have produced a wide range of performance results that are negative or not conclusive (Tam, 1998), mixed (Avison, Eardley, & Powell, 1998; Ranganathan & Samarah, 2001), or positive a positive and significant relationship between IT investment and firm financial performance (Im, Dow, & Grover, 2001). Kamssu, Reithel, & Ziegelmayer (2003) explore the impact of information technology and stock returns. They conclude that Internet-dependent firms have lower excess returns than non-Internet firms do in a booming economy and that Internet stocks trade at relatively higher prices than non-Internet stocks. The explosion of Internet technology and behavior of investors and decision makers toward firms that use the Internet suggest that Internet technology must have an impact on firms' market performance.

Stock performance helps investors gauge how well their managers are handling their money. Several studies have proposed different methods to assess stock performance. Armitage & Jog (1996), Rogerson, (1997), and Clinton & Chen (1998) have used economic value as a measure of performance. The economic value added is obtained by comparing profits with the cost of capital involved in obtaining these profits (Stephens & Bartunek, 1997). Johnson & Pazderka (1993) and Sundaram, John, and Kose (1996) have employed stock market performance estimates to measure firm performance. Fama & French (1995), Loughran (1997), Zaher (1997) and Ranganathan & Samarah (2001) employ the stock excess returns based on the Capital Asset Pricing Model (CAPM) to measure stock performance. Historically, the stock values of information technology firms bear very little relationship to classical business performance measures (Savitz, 1998), which creates a need for non-traditional proxies and estimation methods.

The statistical methodology incorporated in this study employs a nonparametric approach to comparing the stock market performance of firms across a decade of six different development stages for the information technology industry. The study uses multiple years of data based on the diffusion model hypothesis that the spread of information needs time and stock price momentum reflects gradual diffusion of firm-specific information (Hong & Zhu, 2006). There is no research focusing on stock market volatility of computer network and information technology services companies.

TECHNOLOGY ERAS

Between 1996 and 2006, several major events in the field of information technology made a lasting impact on many businesses and consumers. Six implicit periods are identified for the purposes of this study. Although somewhat arbitrary, the six periods are placed in twenty-month segments in an effort to capture stock market returns in a broad representative timeframe. The six period classifications are the browser era, Y2K era, post-Y2K era, post-9/11 era, outsourcing era, and mobile/wireless era.

The browser era is defined in the study as the 20-month period of August 1996 through March 1998. The World Wide Web was but a few years old when Mosaic, often considered the first browser, was introduced. The web was massive and complicated. Prior to Mosaic, access to the Internet was largely limited to text, with any graphics displayed in separate windows. Users needed to possess certain technical knowledge and skills to exploit available capabilities and access both the Internet and the web. Mosaic eventually became Netscape. The success of Netscape gained the attention of Microsoft, which developed the Explorer browser. A cluster of related and supporting technologies came together to make the browser a significant innovation breakthrough. The browser era developed with the assistance of computer servers, bandwidth affordability and availability, content providers, and communication links. The browser interface made it easier for users to connect to the web and created a significant critical mass of users (Cocca, 2005). The use of browsers to connect to the Internet pressured software developers and content providers to adhere to certain accepted specifications and standards. These standards and specifications enhanced the interoperability of web-related products and services. For years, enterprises struggled to find reliable, cost-effective ways to integrate and automate critical processes between different application packages. The web-enabled applications and technology provided the enterprises with the ability to integrate different systems and application types regardless of their platform, operating system, programming language or locations. In essence, the browser was the key that unlocked the World Wide Web to a massive number of users. Netscape was the most used browser to access the web. It allowed millions of users to navigate the web and was the vehicle that linked people and information. The catalyst marked the boom in the Internet. The browser made it possible for millions of users to access the web daily, to send messages and to perform business transactions that would not have been possible without the browser. The browser has changed the way society communicates, created new businesses and contributed to the demise of other businesses.

The Y2K era in this study is the 20-month period of April 1998 through November 1999. In the early days of software development and hardware design, it was common practice to use standard two-digit shorthand to indicate the year. This practice infiltrated many software applications and hardware designs. In the early nineties, this became known as the Y2K problem. The Y2K problem implied that some software and hardware would not perform as expected after December 31, 1999. While many were relieved that the catastrophic consequence of Y2K did not materialize, it is clear that this era had profound impact on the amount of expenditures in the field of information technology. The commercialization of the Internet and the need to overhaul information technology infrastructure in preparation to address the potential Y2K problem was a significant driving force. The Department of Commerce estimated that there was approximately $100 billion spent to address the Y2K problem in the United States (Manion & Evan, 2000). The significance of Y2K is more than the expenditure amount, it also provided opportunities to shift to new computing platforms, implementing new approaches to software applications development and highlighting the relevant role of information technology to the overall enterprise's business strategy.

The post-Y2K era in this study is the 20-month period of December 1999 through July 2001. The 2001 year had been a bust with the dot-com implosion and the downturn of the economy. The pre-Y2K buildup resulted in the post-Y2K bust for many information technology companies. Many companies cut back on information technology expenditures during this era because of the significant expenditures in the preceding era. Despite the bursting of the dot-com bubble, significant advances in information technology advances continued during this era. The importance of critical infrastructure, the need for compliance with security regulations, the importance of business continuations plans, and data mining/warehousing were four major themes that emerged during this time (Terry, Macy, & Abdullat, 2010).

This study defines the post-9/11 era as the 20-month period of August 2001 through March 2003. The event of September 11 accentuated the importance and the vulnerability of information technology in the event of catastrophic attack. It necessitated the need to develop plans to identify its critical infrastructure that is required to maintain minimum operation of the economy and government. The security of critical infrastructure became a vital concern. Security of critical infrastructure and other resources went through extensive change to mitigate the risks. Federal regulations tightened security regulations to include many aspects of business processes and functions. Information technology was targeted as the means to meet the security concern. The sense of urgency to meet security demands and concerns by the federal government made it easier to fund many of the new research and development activities by businesses. Moreover, many businesses recognized the value of computer security as a large, emerging market. During this time, the importance of data centers' redundancy of data and the need for diversity of geographic concentration of information technology resources gained in relevance and significance (Terry, Macy, & Abdullat, 2010). In addition, network infrastructure influenced businesses in a very profound manner that required continued increase in computing power. Barriers that existed between firms for most of the 20th century gave way to accommodate the need for partnership-based opportunities afforded through e-business. The need for interoperability and flexibility increased during this era to exploit new business opportunities. This created a demand for new system architectures to mitigate the shortcomings of grid computing and client server technologies. The continuous decline in the storage cost of data, the increase of computing power, and the availability of broadband bandwidth reduced the incentive for firms to discard any data (Hong & Zhu, 2006). The availability of stored digital data and information presented firms and government agencies with a major challenge to identify ways to make some sense of the huge amount of data. The government's heightened concern with security was instrumental in funding new developments in data mining and contributed to the increased use of business-intelligence software to mine huge amounts of stored data.

The outsourcing era is defined in the study as the 20-month period of April 2003 through November 2004. In this era, companies were looking for different measures to cut costs and to improve the balance sheet. Outsourcing and off shoring became prominent business strategies to reduce operational cost, to enhance services, and to improve financial performance. In addition to the economic and market conditions, three Laws influenced this period: Moore's (growing power of computer chips), Metcalfe's (growing network usefulness) and Gilder's (growing communications bandwidth). These laws transformed processes, products and services (Terry, Macy, & Abdullat, 2010). Combining the economic conditions and the changes in information technology made it possible to reduce cost but to continue performing certain functions of the business at the same or higher level. Businesses quickly realized the cost advantage of developing and maintaining their software applications in India, China and Eastern Europe. In looking back at that era, it is clear that notwithstanding the challenging economic conditions at the time, it marked the beginning of accepting outsourcing as a cost-reduction strategy (Hong & Zhu, 2006). The outsourcing phenomena affected many areas of information technology including software development and programming, technical support, calling centers and customer services.

The mobile/wireless era is the 20-month period of December 2004 to August 2006. The term mobile computing is the use of portable computing devices either in transit or from a remote location. Wireless technology had been around for many years, but the industry-transformed society during this era. The mobile computing environment is composed of small devises that permit users to have access to information almost anywhere at any time (Cocca, 2005). The increased access by users to the Internet, the innovation of wireless technology, and the high number of cellular phone services contributed to the growth of mobile computing. Moreover, the dependency and the reliance on laptops and hand-held devises to perform computing functions increased the demand for mobile and wireless products and services.

INDUSTRY OVERVIEW

The computer network and information technology service industry is characterized as one that must be extremely nimble and fast at meeting the needs of customers. The firms' customers in this industry are primarily large businesses both within the technology sector but also in all other sectors. Customers who purchase network equipment seek equipment that will meet their needs for an extended time period but with the capability of being upgraded as needed. High-end network equipment can cost over $100,000, so the purchases are viewed as capital equipment and are highly scrutinized by the buyers. While price does factor into the decision, buyers also recognize that performance and ease of maintenance are other major, important decision factors. For the technology services part of the industry, customers seek products that come with substantial and timely service packages. For the firms, selling the equipment and then the service package extends the value of each sale and provides an opportunity to solidify the relationship and thus expand it into future sales. Because of the size and technical requirements of the hardware in this industry, it is capital intensive but with a requirement of highly-skilled employees who can explain to the customers how the product can benefit the buyer firm. In order to stay fresh and gain access to new markets, many of the firms are active acquirers of competitors. The goal of the acquisitions is to gain a blockbuster product. Customers and stock investors quickly change allegiance to the latest hit product and reward it with sales and stock price jumps. The five computer network and information technology services companies included in the study are Cisco Systems (CSCO), 3Com (COMS), Ericsson (ERIC), Nortel Networks Corporation (NT), and Yahoo Inc. (YHOO). The five firms in this study all followed an aggressive acquisition strategy to gain access to blockbuster products or customer bases.

Cisco Systems (CSCO) is a leading supplier of internetworking hardware, such as routers and switches, to direct data, including voice and video. Approximately half of its revenues are from outside of North America (Value Line, 2010). Like the other firms in the industry, Cisco is an active buyer of competitors and firms in related but new markets. Notable purchases over the years have included Kalpana, a switch manufacturer in 1994, Percept Software, video transmission software maker in 1998, Andiamo Systems, storage network switch maker in 2002, Linksys, home network specialist in 2003 and WebEx Communications, conference systems in 2009 (Hoover's Online, 2010). The list demonstrates how Cisco uses acquisitions to strategically place itself in all points of the networking supply chain but without having to conduct the initial research & development required of each new sub-area of networking. Cisco's best period was the Y2K period. Firms in all industries were upgrading systems and seeking better control over their entire information technology structure. Earnings per share during the Y2K era grew at over 35% annually (Value Line, 2010). While slightly slower than during the browser era, investors recognized the central role Cisco played in the technology sector and rewarded it with a P/E ratio well above 100 (Business & Company Resource Center, 2010). During the post-Y2K era, earnings dropped by over 50%, even though revenues increased (Value Line, 2010). The firm had invested heavily in Internet protocol network equipment, whose sales dropped sharply during the technology bust (Mergent Online, 2010). The stock price dropped sharply, matching the decline in earnings. The post-9/11 era was tough for Cisco as it looked to diversity its product line and rebuild cash flow. By 2004 during the outsourcing era, Cisco had rebounded enough to resume its acquisition strategy (Hoover's Online, 2010). Stock investors responded positively to Cisco's strategy and increased the P/E ratio to about 30, giving Cisco its second best performing era (Value Line, 2010). Cisco's upward trend continued during the mobile/wireless era, albeit at slower growth rates. Its P/E ratio fell by 1/3 as its stock price trended upward slightly (Value Line, 2010). Overall, Cisco has learned from its mistakes during the post-Y2K era and maintains an acquisition strategy that does not deplete its cash.

3Com (COMS) is the upstart networking firm in the industry who tries to play with the big players. With a market capitalization that usually ranges from 1% to 3% of Cisco's market capitalization, 3Com spends as if it is the biggest player in town including naming Candlestick Park in San Francisco 3Com Park from 1995 to 2000 (Value Line, 2010). Robert Metcalfe, the inventor of the Ethernet for Xerox, founded 3Com. Using the Ethernet as its base technology, 3Com developed ancillary hardware products (Hoover's Database, 2010). Of the five firms in the study, 3Com had the lowest performance in the browser and Y2K eras. Operating costs fluctuated as the firm tried to integrate all of the firms it had acquired (Value Line, 2010). In particular, 3Com purchased U.S. Robotics and the Palm Pilot PDA in 1997. Integrating the larger U.S. Robotics was problematic and resulted in layoffs, inventory problems, negative press, and finally, one of the largest shareholder lawsuits and settlements in history (Hoover's Online, 2010). During the post-Y2K era, the stock price plummeted from above $100 per share to under $10 per share (Value Line, 2010). In an effort to raise cash, 3Com spun-off Palm in 2000 (Business & Company Research Center, 2010). The post-9/11 era saw 3Com refocus its business away from consumers and to business along with reducing 30% of its workforce (Hoover's Online, 2010). Investors responded positively to the firm's actions and rewarded it with an increasing stock price, albeit modestly (Value Line, 2010). This is especially surprising considering that 3Com posted negative earnings from 2001 through 2006. During the outsourcing and mobile/wireless eras, 3Com continued to divest itself of ancillary product lines, which raised cash for the firm (Hoover's Online, 2010). Additionally, the firm began to move aggressively into Asia, particularly China. Its acquisitions and strategic alliances produced new products and sales (Mergent Online, 2010). China is the source of over half of 3Com's sales (Value Line, 2010). These actions helped 3Com return to profitability and a positive cash flow by the end of the wireless era.

Ericsson (ERIC), one of the largest Swedish companies, is a leading provider of telecommunication and data communication systems and related services covering a range of technologies, including especially mobile networks. Directly and through subsidiaries, it also has a major role in mobile devices and cable TV and IPTV systems (Value Line, 2010). Throughout the 1990s, Ericsson held a 35-40% market share of installed cellular telephone systems (Hoover's Online, 2010). Like most of the telecommunications industry, Ericsson suffered heavy losses after the telecommunications crash in the early 2000s. It was forced to do a 1-for-10 reverse stock split in 2002 (Value Line, 2010). On October 1, 2001 the handsets division formed a joint venture with Sony called Sony Ericsson. Ericsson is now a major provider of handset cores and an infrastructure supplier for all major wireless technologies (Hoover's Online, 2010). It has played an important global role in modernizing existing copper lines to offer broadband services and has actively grown a new line of business in the professional services area. Ericsson's focus on the hardware for networks has allowed it to survive the rough times. Its North American business is less than 10% of total sales while Europe is more than 50% of revenues (Value Line, 2010). Ericsson, while considered a quality product, has never been able to make a huge dent in North America because its wireless products are functional but without the features of an iPhone or BlackBerry (Business & Company Resource Center, 2010). In contrast, its network hardware has a strong reputation and is the growth engine for the firm. The firm's net profit margin bounced from negative values in 2001 and 2002 to over 11% by 2004. Its focus on infrastructure hardware is profitable; the net profit margin has been close to or over 15% since 2005 (Mergent Online, 2010). U.S. investors have not recognized fully the strengths of Ericsson's business. The lack of a consumer presence has resulted in a declining P/E ratio. Ericsson's P/E ratio was close to 90 in 2000 but less than 20 since 2004 (Value Line, 2010). It is one of the few technology companies to pay a dividend, which increased yearly since the 2005 reinstatement. Over the entire 120-month period, Ericsson has the lowest total return.

Nortel Networks (NT/NRTLQ) is a leader in telecommunications networking. Originally, a part of Bell Canada, Nortel was a division in a series of telecommunications firms until its current incarnation created during the browser and Y2K eras (Hoover's Online, 2010). As one of the first firms to move into the Internet hardware business, Nortel supplemented its product line with an aggressive acquisition strategy acquiring Bay Networks and Shasta Networks (Business & Company Resource Center, 2010). The acquisitions helped Nortel increase its earnings growth rate to above 30% during the browser era. Investors supported the acquisition strategy and increased the P/E ratio by 60% during the Y2K era (Value Line, 2010). However, Nortel's push into the Internet business ensured that it would suffer when the technology bubble burst. Its earnings per share and cash per share turned negative. Investors punished the stock, whose price fell by 95% during the post-Y2K era (Value Line, 2010). Nortel responded by realigning its business and cutting its workforce. In 2001 alone, it laid off 50,000 employees. It also sold business, sometimes at a loss, to gain cash (Hoover's Online, 2010). By the end of the post-9/11 era, the firm had returned to positive earnings and cash flow (Value Line, 2010). Nortel was an active member of the outsourcing era. It signed a deal with Singapore's Flextronics to outsource all of its manufacturing. This allowed Nortel to focus on design and marketing of products but not the quality control issues associated with manufacturing (Hoover's Online, 2010). Investors responded positively and pushed the stock price above $80 (Value Line, 2010). During the mobile/wireless era, Nortel focused on increasing the speed and capabilities of its network products. In particular, it focused its research and development on 3G and 4G technologies (Mergent Online, 2010). Overall, Nortel has the lowest performance of the five stocks examined. After starting out on a high, Nortel could not recreate the products or the excitement once the Internet became routine and the focus needed to shift to cost-control, which was not a strength of Nortel.

Yahoo! Inc. (YHOO) is a leading provider of Internet services including search, auctions, and mail to customers. Unlike the other firms in this study, Yahoo focuses more, but not completely, on services to the retail consumer (Hoover's Online, 2010). Yahoo went public during the browser era even though it did not have positive earnings until 1998 (Value Line, 2010). Just as the other firms in the industry used acquisitions as a central part of their corporate strategy, Yahoo actively purchased firms with new ideas or existing customers. During the Y2K era, Yahoo looked for firms with products to monetize the Internet such as direct marketing firm Yoyodyne and Internet communications firm Broadcast.com (Hoover's Online, 2010). During the Y2K era, Yahoo's dominance of the search engine and ability to help businesses turn a profit on Internet customers was rewarded by stock investors who made it the darling of the Internet boom and increased its P/E ratio to over 500. However, during the technology bust, Yahoo struggled and its stock price fell to lows not seen since just after it went public (Value Line, 2010). The post-Y2K era was a time of restructuring for Yahoo. In 2000, it announced it would charge fees to list items on its auction site. Users responded by abandoning the site (Hoover's Online, 2010). Yahoo struggled to find a way to make consumers pay for Internet services. Yahoo reduced its workforce by about 1000 employees (Hoover's Online, 2010). During the post-9/11 era, Yahoo moved into new areas including music and ebooks. It also redesigned its webpages to allow for more advertising (Business and Company Resource Center, 2010). Stock investors responded positively and increased the stock price (Value Line, 2010). During the outsourcing era, Yahoo finally regained its stride. Increasing revenue from online advertising and paid search resulted in a doubling of sales and earnings during this period. Yahoo even had a 2-for-1 stock split in 2004 (Value Line, 2010). By the mobile/wireless era, Yahoo continued to expand its reach internationally, primarily Asia, and domestically, targeting Hispanics. It also purchased Flickr, the photo site, to augment its personal pages offerings (Hoover's Online, 2010). Revenue continued to grow but earnings slowed because of the cost of the acquisitions and the resulting integrations. Investors did not overly punish the stock but did decrease the P/E ratio (Value Line, 2010). During the six eras, Yahoo refocused itself into an Internet advertising services firm as it sought the latest blockbuster Internet trend on which to capitalize through advertising. Overall, Yahoo had the highest total performance of all the firms, albeit partially because the price started so low in the browser era.

DATA AND METHODOLOGY

Is there a difference in the stock market performance of computer network and information technology services companies in the different period classifications? In this section, we compare the stock market returns of computer network and information technology services companies in six different twenty-month periods between the years 1996 through 2006. Five different information technology firms specializing in computer network and information technology services are the focus of this study. The primary data source is the Yahoo! finance website, which offers daily and monthly closing stock prices across multiple years. The six period classifications are the browser era, Y2K era, post-Y2K era, post-9/11 era, outsourcing era, and mobile/wireless era. The statistical methodology incorporates a nonparametric approach to comparing the stock market performance of a company in the six different periods. The Kruskal-Wallis test offers the most powerful test statistic in a completely randomized design without assuming a normal distribution. A traditional event study methodology is not applicable to this specific research design because the research periods require a long time horizon instead of the narrow window associated with an event study. In addition, a nonparametric approach is more efficient given the limitation of defining all six periods as strict twenty-month periods given some eras might be somewhat longer or shorter than the twenty-months.

The Kruskal-Wallis test is sensitive to differences among means in the k populations and is extremely useful when the alternative hypothesis is that the k populations do not have identical means. The null hypothesis is that the k company stock returns in the different periods come from an identical distribution function. For a complete description of the Kruskal-Wallis test, see Conover (1980). The specific equations used in the calculations are as follows:

(1) N = [[summation].sub.i][n.sub.i] with i = 1 to k

(2) [R.sub.i] = [[summation].sub.j]R([X.sub.ij]) with j = 1 to [n.sub.i]

(3) [R.sub.j] = [[summation].sub.i][O.sub.ij] [R.sub.i] with i = 1 to c

(4) [S.sup.2] = [1/(N - 1)] [[[summation].sub.i] [t.sub.i] [R.sub.i.sup.2] - N[(N + 1).sup.2]/4] with i = 1 to c

(5) T = (1/[S.sup.2]) [[[summation].sub.i]([R.sub.i.sup.2]/[n.sub.i]) - N[(N + 1).sup.2]/4] with i = 1 to k

(6) [absolute value of ([R.sub.i]/[n.sub.i]) - ([R.sub.j]/[n.sub.j])] > [t.sub.1-a/2] [[[S.sup.2](N - 1 - T)/ (N - k)].sup.1/2] [[(1/[n.sub.i]) + (1/[n.sub.j])].sup.1/2]

where R is the variable rank and N is the total number of observations. The first three equations find average ranks. Equation (4) calculates the sample variance, while equation (5) represents the test statistic. If, and only if, the decision is to reject the null hypothesis, equation (6) determines multiple comparisons of stock market returns across the various periods.

RESULTS

Table 1 offers summary statistics for the five computer network and information technology services companies in the research cohort. Yahoo is the most volatile company in the research sample with the largest mean, median standard deviation, sample variance, and maximum monthly return. Nortel is the sample representative with the minimum monthly return of greatest magnitude. Monthly returns for the companies range from a minimum of -0.5478 for Nortel to a maximum of 1.3365 (or 13% in one month) for Yahoo. Ericsson is the median firm for five of the seven categorical descriptive statistics. The most notable observations are the very large 120-month return of 3,416% for Yahoo and the respectable 120-month return of 275% for Cisco. Three of the five companies in the research cohort earn 120-month returns that are negative or relatively small, with Nortel Networks earning -66%, 3Com earning -56%, and Ericson earning a modest 23%. The negative returns earned by Nortel and 3Com help explain the reason for the bankruptcy and sell off of Nortel in 2009 and the 2010 acquisition of 3Com by Hewlett-Packard.

The nonparametric empirical approach yields four T-values of 27.23 (p-value = .0001) or higher, indicating a significant difference in stock market returns across the six period classifications for all companies in the study. Table 2 presents a summary of the average rank value of stock market returns for each company across the six periods defined in this study. Assuming an alpha level of .05, the empirical results from equation 6 indicate all companies have four or more time-periods with stock market returns that are statistically different. The most interesting observation from Table 2 is the low relative return earned in the post-9/11 (period 4) era. Four of the five companies achieve their lowest return period in the post-Y2K or post-9/11 eras. The results imply companies in the same industry all tend to face financial challenges during the declining phase of a stock market bubble. The only company that deviates from the post-Y2K and post-9/11 negative trends is 3Com, which achieves their low return period in the Y2k era and achieves a high return period in the post-9/11 era. Although the relatively consistent negative return in the bubble bursting eras might seem obvious, it is important to note all the companies in the study survived the stock market bubbles of the postY2K and post-9/11 eras. One of the limitations of the study is a potential survivor firm bias, where companies that did not survive the stock market bubble burst of the post-Y2K or post-9/11 eras are not part of the study. This limitation is somewhat mitigated by the observation that companies that did not survive almost certainly hit low periods in the post-Y2K or post-9/11 eras, which is consistent with our empirical results. The fact that even survivors consistently struggled and only 3Com prospered is noteworthy given recent acquisition of 3Com by Hewlett-Packard.

The high return period for computer network and information technology services companies is more diverse than the low return period. Four of the five companies achieve their high return period in different eras, which demonstrations a high degree of performance differential across firms in the industry during a bull market. Nortel Networks achieves a high return period in the browser era. Cisco and Yahoo achieve their high return period in the Y2K era. The high return period for 3Com is the post-9/11 era. The high return period for Ericson is the outsourcing era. The variation in the high return periods across the five companies provides evidence the computer network and information technology services industry produces blockbusters. Items that are blockbusters tend to have one product or innovation that captures the attention of investors. The product does not have to be the most profitable item but investors normally consider the innovation to have strong potential for success. Industries characterized as containing blockbusters normally have low correlation with respect to price and stock market returns because product innovation is sporadic across the industry.

CONCLUDING COMMENTS

The purpose of this research is to compare the stock market performance of five companies specializing in computer network and information technology services across six information technology eras. The six period classifications are the browser era, Y2K era, postY2K era, post-9/11 era, outsourcing era, and mobile/wireless era. The statistical methodology incorporates a nonparametric Kruskal-Wallis test to compare the stock market performance of the companies in the research cohort. The primary data source is the Yahoo! finance website.

The results of this study imply a high correlation of stock market prices for computer network and information technology services companies during bear markets but a low degree of correlation with respect to firms achieving their peak return period. Specifically, four of the five companies achieve their peak return period in different eras. The variation in the high return periods across the five companies provides evidence the computer network and information technology services industry produces blockbusters.

One of the limitations of the study is a potential survivor firm bias, where companies that did not survive the stock market bubble burst of the post-Y2K or post-9/11 eras are not part of the study. This limitation is somewhat mitigated by the observation that companies that did not survive almost certainly hit low periods in the post-Y2K or post-9/11 eras. A second limitation of the study is the application of stock market returns across a very broad timeframe encompassing 120 months. Traditional finance event studies usually focus on daily data for a very short window of time in order to minimize the potential contamination of other events. This study requires the use of a larger than normal research window in order to compare the six different period classifications. Thus, the results should be interpreted with caution given the potential for correlation with other events that occurred in any given focus era. One avenue for future research is to examine consistency of the empirical results across various eras by employing multiple short-run event studies.

REFERENCES

Allen, F. & Gale, D. (2000). Financial contagion. Journal of Political Economy, 108(1), 1-33.

Armitage, H. & Jog, V. (1996). Economic value creation: What every management accountant should know. CMA--The Management Accounting Magazine, 70(8), 21-24.

Avison, D. & Eardley, W. (1998). Suggestions for capturing corporate vision in strategic information systems. Omega, 26(4), 443-459.

Bartholomew, D. (1998). How to lead? Industry Week, 247(2), 27-30.

Business & Company Resource Center (2010). Company Financials on Electronic Arts, Hewlett-Packard Compaq, IBM, Microsoft, Oracle, and Sun Microsystems. New York: Gale Cengage Learning. Retrieved March 12, 2010.

Camerer, C. (1989). Bubbles and Fads in Asset Markets: A Review of Theory and Evidence. Journal of Economic Surveys, 3(1), 3-38.

Clinton, B. & Chen, S. (1998). Do new performance measures measure up? Management Accounting, 80(4), 38-42.

Cocca, T. (April 2005). What made the Internet bubble burst? A butterfly flapping its wings or how little things can make a big difference. Available at SSRN: http://ssrn.com/abstract=758346.

Conover, W. (1980). Practical Nonparametric Statistics. New York: John Wiley & Sons Publishing.

Demers, E. & Lev, B. (2001). A rude awakening: Internet shakeout in 2000. Review of Accounting Studies, 6(2), 331.359.

Fama, E & French, K (1995). Size and book-to-market factors in earnings and returns. Journal of Finance, 50(1), 131-155.

Higgins, M. & Osler, C. (1997). Asset market hangovers and economic growth: The OECD during 1984-1993. Oxford Review of Economic Policy, 13(3), 110-134.

Hong, W. & Zhu, K. (2006). Migrating to Internet-based e-commerce: Factors affecting e-commerce adoption and migration at the firm level. Information & Management, 43(1), 204-221.

Hoover's, Inc. (2010). Company records on Electronic Arts, Hewlett-Packard Compaq, IBM, Microsoft, Oracle, and Sun Microsystems. Hoover's Company Records. Austin, Texas: Hoover's Company Reports. Retrieved March 12, 2010.

Im, K., Dow, K. & Grover, V. (2001). A reexamination of IT investment and the market value of the firm--An event study methodology. Information Systems Research, 12(1), 103-117.

Johnson, L. & Pazderka, B. (1993). Firm value and investment in R&D. Managerial and Decision Economics, 14(1), 15-24.

Kamssu, A., Reithel, B., & Ziegelmayer, J. (2003). Information technology and financial performance: The impact of being an Internet-dependent firm on stock returns. Information Systems Frontiers, 5(3), 279-288.

Loughran, T. (1997). Book-to-market across firm size, exchange, and seasonality: Is there an effect? Journal of Financial and Quantitative Analysis, 32(3), 249-268.

Manion, M. & Evan, W. (2000). The Y2K problem and professional responsibility: A retrospective analysis. Technology and Society, 22(3), 361-387.

Mergent Online (2010). Company profiles on Electronic Arts, Hewlett-Packard Compaq, IBM, Microsoft, Oracle, and Sun Microsystems. Retrieved March 11, 2010.

Molloy, S. & Schwenk, C. (1995). The effects of information technology on strategic decision making. Journal of Management Studies, 32(3), 283-311.

Ofek, E. & Richardson, M. (2003). Dotcom mania: The rise and fall of Internet stock prices. The Journal of Finance, 58(3), 1113-1138.

Ranganathan, C. & Samarah, I. (2001). Enterprise resource planning systems and firm value: An event study analysis. 22nd International Conference on Information Systems, 157-158.

Rogerson, W. (1997). Intertemporal cost allocation and managerial investment incentives: A theory explaining the use of economic value added as a performance measure. Journal of Political Economy, 105(4), 770-795.

Sampler, J. (1998). Redefining industry structure for the information age. Strategic Management Journal, 19(4), 343-355.

Savitz, E. (April 20, 1998). A momentary lapse for Internet stocks, but mostly they defy gravity. Barron's, Chicopee.

Sornette, D. & Malevergne, Y. (2001). From rational bubbles to crashes. Physica, 299, 40-59.

Stephens, K., & Bartunek, R. (1997). What is economic value added? A practitioner's view. Business Credit, 99(4), 39-42.

Sundaram, A., John, T. & Kose, J. (1996). An empirical analysis of strategic competition and firm value: The case of R&D competition. Journal of Financial Economics, 40(3), 459-486.

Tam, K. (March, 1998). The impact of information technology investments on firm performance and evaluation: Evidence from newly industrialized economies. Information Systems Research, 9(1), 85-98.

Terry, N., Macy, A, & Abdullat, A. (2010). Stock market volatility: A comparison of vertically integrated technology companies. The Journal of the American Academy of Business, Cambridge, 15(2), 242-249.

Value Line (2010). Company stock reports on Electronic Arts, Hewlett-Packard Compaq, IBM, Microsoft, Oracle, and Sun Microsystems. Value Line Investment Surveys. New York: Value Line, Inc. Retrieved March 12, 2010.

Zaher, T. (1997). The cross-section of expected stock return: Application of the Fama-French methodology to the utilities industry. Journal of Energy Finance and Development, 2(2), 229-238.

Anne Macy, West Texas A&M University

Neil Terry, West Texas A&M University

Jean Walker, West Texas A&M University

Table 1
Summary Statistics for Computer network and
Information Technology Services
Firms Average Monthly Returns

Firm    Mean    Median    Standard     Sample    Minimum   Maximum
                          Deviation   Variance

COMS   0.0112   -0.0070     0.1944     0.0378    -0.5069   0.9300
CSCO   0.0198   0.0252      0.1312     0.0172    -0.3673   0.3892
ERIC   0.0162   0.0015      0.2025     0.0410    -0.5436   1.0273
NT     0.0151   0.0077      0.2320     0.0538    -0.5478   1.2778
YHOO   0.0538   0.2176      0.2384     0.0568    -0.3623   1.3365

Firm   120-month
         Return

COMS       -56%
CSCO       275%
ERIC        23%
NT         -66%
YHOO     3,416%

Notes: The sample period is the 120-months between August 1996
and August 2006. Total return for ten-year period is 102.6% for
the Dow Jones Industrial Average and 91.3% for the NASDAQ
Composite Index.

Table 2
Computer Network and Information Technology services Firms
(Average Rank Order Value of Returns)

       T-values       Period 1     Period 2      Period 3
Firm   (p-value)      8/96-3/98    4/98-11/99    12/99-7/01

COMS   27.23 (.01)    53.5 *       43.8 -        57.7 *
CSCO   30.43 (.001)   80.4 **      99 3 ***      37.0 *
ERIC   35.37 (.01)    90.6 **      58.0 *        49.2 *
NT     33.72 (.01)    95 9 ***     67.4 **       43.3 *
YHOO   39.62 (.01)    82.3 **      97 1 ***      10.8 -

       Period 4     Period 5      Period 6
Firm   8/01-3/03    4/03-11/04    12/04--8/06

COMS   81.3 ***     71.2 **       55.7 *
CSCO   24.5 -       82.5 **       39.3 *
ERIC   17.2 -       103.8 ***     51.2 *
NT     25.2 -       92.5 ***      38.8 *
YHOO   50.9 *       80.2 **       41.4 *

Notes: The first column is a listing of the ticker symbols for the
five computer network and information technology services companies
included in the study. The second column is the value of the
equation (5) test statistic and p-value for each company, which
determines if there is a statistical difference in stock market
returns across the six periods. Columns three through eight present
the average rank value of the stock market returns for the six
periods of the study. Asterisk(*) and negative signs (-) signify
difference in average rank values as follows:

*** Indicates period with highest statistically significant return
derived from equation 6.

** Indicates period with second highest statistically significant
return derived from equation 6.

* Indicates period with third highest statistically significant
return derived from equation 6. - Indicates period with lowest
statistically significant return derived from equation 6.

Some periods do not have a return that is statistically significant
from an alternative period.