Distributed Work - Book Review

espite the increasing prevalence of organizations staffing important projects with team members from across the nation and around the globe, academic research on the effectiveness of these arrangements has remained relatively sparse. Many of us recall Thomas Allen's (1977) Managing the Flow of Technology, which alerted us to how the spatial layouts of workplaces can influence who interacts with whom, and, as a result, can affect employee and team productivity. His account emphasized the importance of locating together those who need to interact with one another. Following this auspicious beginning, however, interest in extending our understanding of space in the workplace appears to have remained rather flat among organizational researchers in the quarter century following the publication of Allen's book. In the meantime, companies have become more geographically dispersed as they globalize and as distant firms merge. Attempting to make the most of their human capital, these firms increasingly assign projects to teams of employees working in different offices around the world. Given Allen's findings, one might worry that these arrangements needlessly hamper the efficiency of global organizations. But the last two decades have also witnessed dramatic improvements in the technologies available for collaborating across distances. Perhaps video conferencing, electronic mail, and instant messaging adequately substitute for co-location.

Advertisement

To reinvigorate this topic, Pamela Hinds and Sara Kiesler have assembled 40 contributors across 18 chapters in Distributed Work. Each of these chapters investigates some aspect of managing teams of geographically dispersed workers. Relative to many edited volumes, Hinds and Kiesler's book gathers an unusually cohesive set of writers--virtually every chapter approaches the issue with the lens of a psychologist. What do they conclude? In 25 years, little has changed. Propinquity remains critical to collaboration; trying to coordinate team members across distances can lead to a variety of problems that these authors usefully ground in psychological theory. Furthermore, the various electronic media proposed as solutions all clearly fail to substitute for face-to-face contact.

Several audiences can appreciate different dimensions of this volume. Managers interested in how to organize project teams will likely appreciate the clear writing, lack of technical detail, and plethora of stories, though the scarcity of specific recommendations may frustrate them. Researchers can harvest many fruitful topics for study by reviewing the dozens of untested but theoretically motivated propositions offered across the chapters. Though the editors see the latter as their audience, the volume may prove most useful as a source of teaching material for college courses, a fertile environment for the conceptual pieces that dominate the volume.

Commenting on the specific chapters in any edited volume forces tradeoffs. Instead of briefly mentioning every chapter, I prefer to highlight two pieces that I found particularly enjoyable. In one of the few pieces reporting the results of an experiment, Judith Olson, Stephanie Teasley, Lisa Covi, and Gary Olson investigated the effect of "radical co-location" on team dynamics and performance. By radical co-location, they mean having all team members working in a single room. Though such intimate quarters may strike many of us as too close for comfort, this study found that these configurations doubled the productivity of the product development teams assigned to them. The researchers attributed this effect to improvements in the frequency and depth of communication among members of the team. Moreover, the familiar setting seemed to grow on people; although before taking part in the experiment, participants said they preferred cubicles, afterwards their preference had shifted to working in a common team room. Olson and her colleagues, thus, interestingly raise the question of whether firms should move in the other direction: instead of assembling workers across multiple locations, perhaps they should gather team members in one location and reconfigure their office spaces to allow them to work in a single room.

Another intriguing contribution appears in the second chapter, in which Michael O'Leary, Wanda Orlikowski, and JoAnne Yates remind us that the last century does not hold a monopoly on the coordination of work across distances. O'Leary, Orlikowski, and Yates examine the organizational practices of the Hudson's Bay Company (HBC) from 1670 to 1826. From its headquarters in London, the HBC managed a far-flung network of outposts across what is now Canada. Through a combination of selective recruitment, socialization, and empowerment, the firm conquered distance without the benefit of air travel or e-mail by fostering an organizational culture that mitigated the problems inherent in managing these distant stations. The authors also helpfully suggest how modern managers might learn from HBC's example.

The Paradox of Empowerment: Suspended Power and the Possibility of Resistance - Book Review

If we pause for a moment and consider the recent proliferation of material that takes empowerment as its subject matter, we are confronted with a bewildering choice of approaches. From starry-eyed proselytizers to hard-nosed critics, from devoted disciples to cynical nay-sayers, it seems that everyone has to take a position on empowerment now. Indeed, empowerment is bandied about so promiscuously these days that it is in danger of suffering from what the historian Edward Peters (1996) called "semantic atrophy"--i.e., a word gets used so often and in so many inappropriate situations that it tends to lose its usefulness, even when it could be most helpful. Because empowerment is such a prime candidate for this kind of rhetorical debasement, one might think that the world is crying out for a book that cuts through the mire to come up with a parsimonious definition so that we can all systematically work through its analytical implications. Wendt has taken up this challenge with alacrity, but, in doing so, he has turned it on its head by showing that developing one stable and universal definition of empowerment is probably impossible and certainly undesirable. In fact, there are any number of possible definitions of empowerment that are, to a greater or lesser extent, founded on some kind of paradox, hence the title of his book.

Advertisement

Wendt has come at his subject matter with one important presupposition: regardless of its specific context or usage, the term "empowerment" is frequently deployed ironically to legitimate management practices that are palpably unempowering to those on the receiving end. In this sense, Wendt is primarily dealing with matters related to the "discourses" of business and management. This interest in the semantics of empowerment is hardly surprising. Wendt is a professor of communications who describes himself on page 2 as a "critical/postmodern ethnographer." For many people, such a term would be sufficient to make them roll their eyes and sigh, "Not another postmodernist." It would be a shame, however, if the reader never got beyond this early confession, for parts of Wendt's book have considerable merit. To be sure, all of it is written in a style that, at times, appears to be willfully dense and obscurantist, but at other times Wendt captures significant points in such a graphic and compelling way that things suddenly click into place and, for a moment, you almost forgive him even his more lurid prose. Unfortunately, most of these "Uh huh" moments are unevenly distributed throughout the book. This really is a book of two distinct halves, for although it is divided into four parts, parts 1 and 2 contain most of the material likely to resonate with an audience that also reads the Administrative Science Quarterly. To be frank, some passages from parts 3 and 4 are likely to try the patience of even the most sympathetic readers, regardless of their disciplinary background. So, be warned: to get something out of this book, be prepared to do some hard work!

The book starts with an introduction that provides the rationale for the main text. The line of argumentation runs as follows: (1) things like "empowerment" and "participation" are making today's organizations better places to work, or so we are told; (2) using postmodern social theory (especially the work of Michel Foucault), we can disabuse ourselves of this myth; and (3) we need to develop innovative strategies--the so-called "possibilities of resistance"--that counter the organizational arrangements associated with this spurious notion of empowerment. In the first half of the book, points (1) and (2) are well made, but Wendt's consideration of point (3) in the second half of the book is less effective.

In part 1, Wendt maps out what he means by an organizational paradox: "... a self-referential and contradictory statement--usually an injunction or imperative--that causes confusion, frustration, silence, and/or a sense of incompetence in the listener, and usually leads to a no-win or double binding situation" (p. 15). His concept of the organizational double bind is particularly important because it goes to the very heart of his critique of empowerment. It is that genuine resistance entails going beyond the feelings of inevitability and powerlessness associated with the paradoxical demands placed on us in today's organizations, stepping outside of existing power relations in some way to bring forth new ones that (one hopes) will be more edifying for the participants. Wendt then goes on to demonstrate how specific organizational practices that invoke empowerment are paradoxical and lead to double binds, in the process taking in such things as participative management, kaizen, total quality management, and teamwork. These chapters should be disconcerting to those who believe that such practices do not have a darker side, though Wendt's use of Zen proverbs or "koans" to introduce each paradox may strike some readers as an unnecessary stylistic conceit.

Headhunters: Matchmaking in the Labor Market - Book Review

Finlay and Coverdill provide an engaging and intriguing glimpse into the relationships among headhunting firms, their client firms, and prospective job candidates. Their study focuses on contingency headhunters--those who receive a fee that is contingent on successfully locating a candidate who is hired--and their client firms within a major metropolitan area in the southeastern United States. The authors make use of a rich, diverse archive of data they gathered through semistructured interviews with headhunters and their clients; fieldwork at various headhunting firms; attending industry seminars and conferences; analyzing industry newsletters and various audio, video, and written training materials; and a mail survey they conducted of recruiting firms belonging to a statewide personnel services association.

Advertisement

What emerges is a fascinating portrait of how headhunters--despite their vulnerable and highly competitive role as labor market intermediaries--deftly exploit information, contacts, emotion, and various interpersonal stratagems to lubricate transactions between hiring firms and job candidates. A particularly distinctive facet of the headhunter's role is the need to manage impressions and close the sale on both sides of the transaction: because the most attractive candidates for any given position are usually the individuals least likely to be searching aggressively for work (because they are highly valued by their current employer), headhunters must not only secure assignments from hiring firms but persuade reluctant candidates to consider "jumping ship" from their current firm. Finlay and Coverdill describe a menu of tactics that headhunters employ toward that end, encouraging candidates to view their current employer cynically, in purely instrumental terms, and focusing prospective hires on "wounds" or aspects of their current employment situation that are "hot button issues" that the headhunter can exploit. They also describe how headhunters must often circumvent resistance from human resource (HR) professionals and line managers in their client firms, who view headhunters as intruding on their prerogatives in hiring employees.

The authors' fieldwork leads them to conclude that headhunters often perpetuate stereotypes with respect to age and physical appearance. Their findings with respect to gender and race are more ambiguous. They find little evidence of explicit attention by headhunters to gender and race (except in engagements targeted specifically as "diversity searches"), but the authors suggest that the premium placed by their clients on social similarity, combined with the powerful incentives for headhunters not to rock the boat, can perpetuate gender and racial inequalities. Thus, reliance by firms on labor market intermediaries doesn't necessarily reduce discrimination and, in fact, may sometimes have the opposite effect.

Consistent with network perspectives on brokerage, Finlay and Coverdill find that headhunters often seek to minimize direct interactions between clients and potential candidates, monopolizing information and manipulating perceptions on both sides. Interestingly, their analyses suggest that such brokerage opportunities actually increase as the stakes of the parties in the transaction increase. One might surmise that both client firms and prospective candidates would have ample reason to confront one another directly, given the high stakes each party (especially the candidate) has in the outcome. Yet Finlay and Coverdill argue that it is precisely because the stakes are so high that intermediaries can play such a valuable role--preserving confidentiality, managing expectations, and buffering the parties from the social, emotional, reputational, and other risks associated with courtship.

Although headhunters seek to build reputations with client firms and prospective candidates, there seem to be few enduring sources of trust on which to build lasting relationships, at least based on economic considerations. The business is fiercely competitive. Clients often misrepresent matters to headhunters (e.g., claim that a search is exclusive when in fact multiple headhunting firms have been engaged). It is difficult for headhunters to create and sustain enduring partnerships with clients (e.g., due to resistance by HR personnel and frequent departures of managers at hiring firms), and because the contingent-fee headhunter is only paid after successfully placing a candidate, client firms often have an incentive to engage multiple headhunters. Given how infrequently candidates change positions and the existence of multiple headhunters, candidates presumably also may be tempted to embellish the truth when dealing with headhunters. And everyone involved knows all of this.

The authors make sense of this puzzle by emphasizing that (1) hiring is a social activity; (2) headhunters often acquire invaluable information by virtue of their social connections with clients and candidates; and (3) it is precisely the emotion and potential embarrassment involved in the hiring process that creates a role for skilled third parties to mediate interactions between the parties. Hiring managers and job candidates may know their own preferences, intentions, and aptitudes well, and they certainly have high stakes in the outcomes. But headhunters broker transactions between these two constituencies on a daily basis (and hope to continue to do so in the future), which ostensibly gives them greater ability and incentive to be dispassionate than hiring firms and candidates might be on their own. Moreover, by controlling the information that each party receives about the other, headhunters are able to create more psychological "bang for the buck" than might obtain if the parties negotiated all the details directly. A key role for the headhunter is not just filling job openings with competent people in a timely manner, but doing so in a manner that leaves both the employer and candidate content, committed, and copasetic. Finlay and Coverdill also argue that by running interference between parties, headhunters are often able to preserve enduring long-term relationships of value to their clients. For instance, client firms are often tempted to recruit personnel from their customers, suppliers, and business partners; by delegating their searches to a headhunter, a client firm can often minimize the fallout that would occur by directly poaching talent from organizations with whom it has key long-term relationships.

Big Steel: The First Century of the United States Steel Corporation, 1901-2001 - Book Review

We live in an age obsessed with the birth of new industries, new companies, and new business strategies. At the same time, the legacies of past decisions, old companies, and mature industries continue to govern much of our economy, our political discourse, and our future. Nearly 102 years after its formation on April Fool's Day in 1901, the United States Steel Corporation continues to exercise power that belies the fact that during its lifetime it has gone from beginning "the century with two-thirds of the nation's raw steel-making capacity to ending it with scarcely more than 10 percent" (p. 355). Most recently, U.S. Steel's continued leadership position as America's largest producer saw it lead a partially successful tariff fight that culminated in a March 5, 2002 Bush administration announcement of tariffs on many foreign steel products entering the United States. As one of the few corporate bridges between the day of Andrew Carnegie and J.P. Morgan and the days of Bill Gates and Michael Dell, a history of U.S. Steel's turbulent century is an intriguing topic for examination. Kenneth Warren's book represents an attempt to document a remarkable transition in American industrial history and economics.

Advertisement

Warren's book is not an "easy read." The pages are lathered with voluminous quotations of commodity prices, names of long-dead individuals and companies, and accounts of corporate decisions long buried in internal corporate archives. For a patient reader, however, the history of U.S. Steel raises a host of mysteries of economics and business strategy. Warren's history also represents a case study of a non-monopoly company that, for much of its history, acted as if it were a monopoly. Even the birth of U.S. Steel is laden with more than normal historical significance, even for a company of its size. The formation of U.S. Steel by a group led by J.P. Morgan, Charles M. Schwab, and Elbert H. Gary (after whom Gary, Indiana is named) was in large part a buyout of Carnegie that represented his retirement from the steel business. Carnegie's departure by itself altered the strategic behavior of companies in the industry. Carnegie's spirit of "competition to the death" was replaced by U.S. Steel's desire to avoid "destructive" competition. In addition, the formation of U.S. Steel allowed Carnegie to "monetize" his fortune and to devote his very active retirement to his many charitable endeavors.

U.S. Steel was founded with the hope that its size would lead to economic benefits in the form of market power and operating and network efficiencies. While some cost savings were achieved, and some of the superior management of the Carnegie companies did transfer to the new entity, in the grand scheme of things, size appears to have acted as a drag on the new company rather than as an advantage. Warren hints at some of this tension when he notes that those "who had worked at Carnegie found it difficult to work in reasonable amity with rival companies rather than competing ruthlessly with them as in the past" (p. 27). In the face of attempts to manage prices and exercise leadership, U.S. Steel saw its many small rivals eat away at its markets. Between 1901 and 1927, U.S. Steel's market share in raw steel dropped from 65.7 percent to 41.1 percent (p. 129). This period of U.S. Steel's relative competitive latency allowed for the growth of companies such as Bethlehem Steel, which by 1903 was run by U.S. Steel "defector" Charles M. Schwab. U.S. Steel's price leadership strategies in the early twentieth century may have led to a high return on sales, but the company's dollar sales were essentially stagnant despite significant additions to capacity and production.

By 1936, the stagnation at U.S. Steel was such that Fortune magazine "recalled that the Corporation's policy had once been summarized as 'No inventions: no innovations'" (p. 132) and Charles M. Schwab reported that "the chairman of US Steel admitted to him that the Corporation, in fact, had missed every 'new thing' in steel" (p. 132). Under Carnegie's reign, Pittsburgh-based steel facilities had competed successfully against growing location-based advantages of other regions by relying on innovation, efficiency, and superior management. U.S. Steel under Gary did participate in the geographic dispersion of American steel-making but its "Pittsburgh Plus" pricing (an artificial attempt to exercise industry price discipline by linking prices across America to steel prices in Pittsburgh plus freight from Pittsburgh) led to a hobbling of the corporation's growth into new markets and eventually shrank the region in which Pittsburgh-area steel was competitive. Even in its infancy, U.S. Steel was an illustration of inertia and captivation by sunk-cost investments.

Warren effectively documents U.S. Steel's triumphs and failures over the course of a century. We learn of U.S. Steel's triumphs, such as the fortuitous or prescient refinancing that allowed it later to weather the Depression, and its contributions to the country's war efforts. The impressive gains the company made in the early 1950s led the Economist and Fortune magazine to gush over the company's performance, size (comparable to the production of the Soviet Union or that of Britain and Germany combined), and its transformation from a "laggard" (p. 223). Warren is also forced to describe U.S. Steel's history of tenuous labor relations, its failure to adopt oxygen converters in the 1950s, its falling competitiveness relative to foreign producers in the 1960s, and its growing competitive disadvantages relative to rising "mini-mill" production from the 1960s on.

The ripple effect: emotional contagion and its influence on group behavior

Understanding shared social processes in groups is becoming increasingly important as firms move toward a greater team orientation. These shared social processes can serve as a conduit for a variety of group interactions and dynamics important to getting work done. Interestingly, research on the influence of shared social processes has focused almost exclusively on its cognitive aspects--how ideas and cognition are shared among group members. This can be seen in the social-information processing literature, which focuses on how people are influenced by the cognitions and attitudes of others in their social environment (e.g., Salancik and Pfeifer, 1978; Bateman, Griffin, and Rubinstein, 1987; Shetzer, 1993), as well as in research examining shared social cognitions, which also focuses exclusively on the process through which people construct and share thoughts, ideas, and memories (e.g., Moreland, Argote, and Krishnan, 1996; Cannon-Bowers and Salas, 2001).

While understanding how people share ideas adds to the knowledge of group dynamics, it does not give a complete picture. One also needs to take into account the sharing of emotions, or emotional contagion, that occurs in groups. The importance of emotions in organizational behavior, particularly at the individual level, has been solidly established (see Brief and Weiss, 2002, for a review), and researchers have begun to turn their attention toward understanding the processes and outcomes of collective emotion (see Barsade and Gibson, 1998; Kelly and Barsade, 2001; George, 2002, for reviews). Some theorists have gone so far as to say that "feelings may be the way group entities are known" (Sande-lands and St. Clair, 1993: 445) and that the development of group emotion is what defines a group and distinguishes it from merely a collection of individuals.

Advertisement

Implicit attention has been paid to collective emotion in the organizational behavior literature, with many organizational processes grounded in such affective relations of group members as morale, cohesion, and rapport (Tickle-Degnen and Rosenthal, 1987). The advancement of the emotions literature in psychology has also allowed for a more focused and explicit examination of collective emotion. George and colleagues showed that not only do group emotions exist, but these emotions, which they call group affective tone, can influence work outcomes (George, 1989, 1990; George and Brief, 1992). In a study of senior management teams, Barsade et al. (2000) found that a group's affective diversity, another way of conceptualizing group emotion, also had an effect on individual attitudes and team dynamics. But the question remains, what is the process by which these effects occur?

While literature on shared cognitions can provide some insight into how collective emotions occur via emotional contagion, there are some important differences between emotional and cognitive contagion. First, the transfer of ideas is qualitatively different from the transfer of feelings. Words are central to understanding ideas yet are least important in understanding emotions, for which nonverbal cues are primary (Mehrabian, 1972). Because of the importance of these nonverbal cues, direct interpersonal contact is important for the transmission of emotions in groups. Conversely, sharing cognitions need not occur face to face (Ilgen and Klein, 1988). There are also some differences in the amount of effortful processing involved in cognitive and emotional contagion. Although emotional contagion can contain elements of purposeful processing found in cognitive contagion--such as the evaluation, interpretation, expectation, and personal goals found in the sharing of ideas (Salancik and Pfeifer, 1978)--emotional contagion research studies show that emotional contagion most often occurs at a significantly less conscious level, based on automatic processes and physiological responses (e.g., Hatfield, Cacioppo, and Rapson, 1994; Neumann and Strack, 2000).

Organizational and psychological researchers have begun to investigate the question of emotional contagion through field studies examining mood convergence in work teams. In a field setting, Totterdell et al. (1998) found evidence that the moods of teams of nurses and accountants were related to each other even after controlling for shared work problems. Totterdell (2000) found the same results in professional cricket teams, controlling for the team's status in the game. In a study of meetings of 70 very diverse work groups, Bartel and Saavedra (2000) also found evidence of mood convergence. Similar to Totterdell and colleagues, Bartel and Saavedra showed that work-group mood is something that can be recognized and reliably measured by members in the work group, as well as by observers external to the group. Barrel and Saavedra also examined antecedents to the mood convergence processes and found positive relationships between mood convergence and stable membership in the group, norms about mood regulation in the group, and task and social interdependence. In Totterdell's studies, being older, along with a complex of factors related to being interdependent and satisfied with the team (i.e., more committed to the team, perceiving a better team climate, being happier and engaging in collective activity) were antecedents to mood congruence.

Learning from complexity: effects of prior accidents and incidents on airlines' learning

For all the scientific pizzazz [involved in airline accident investigations], unraveling the subtle, complex chain of events leading to aviation deaths is proving more elusive than ever.

--"Why more plane-crash probes end in doubt," Wall Street Journal, March 22, 1999

Organizations like airlines try to learn from experience, understanding what went wrong so that it won't go wrong next time. But if, as the quote above suggests, the causes are often left in doubt, such learning is likely to be difficult. Learning is also likely to vary across firms, despite industry regulation that should affect all airlines equally. Investigators of the 2000 Air France Concorde crash discovered that British Airways had recommended changes to the Concorde's water deflector in 1995 but that Air France had not made those changes (Phillips, 2000). As Donoghue (1998: 36) explained, "... any safety initiative has an unequal effect on the carriers and becomes an issue to be promoted or fought ... seeking the path that best suits [the airline] individually." Other heavily regulated industries, such as nuclear power, also show substantial variance in incident rates among firms (Morris and Engelken, 1973), which indicates that firms vary in how effectively they learn from their experience. Despite much discussion and analysis of aviation errors (airline accidents and incidents), there has been little work investigating the role of organizational learning and none examining variation in learning across firms in the industry.

Advertisement

Learning from experience has been shown to have important effects on such varied outcomes as manufacturing plant productivity (e.g., Argote, Beckman, and Epple, 1990), service timeliness (Argote and Darr, 2000), and hotel survival (Baum and Ingram, 1998). If firms learn from experience, then the attributes of this experience are likely to affect the rate and effectiveness of learning. Some firms have heterogeneous experience in that their accidents and incidents ("errors") are caused by a large number of different factors, which are likely to interact in complex ways. Some firms have more homogeneous experience, with errors caused by a small number of similar factors. It is likely that the complexity of prior experiences, as well as characteristics of the firms themselves, affect how well airlines can learn from that experience. We investigate these issues in the context of airline accidents and incidents to explain variation in learning among firms in the airline industry.

According to the NTSB (2001) Code of Federal Regulations (49CFR830.2, p. 1195), an accident "means an occurrence ... in which any person suffers death or serious injury, or in which the aircraft received substantial damage." An incident is "an occurrence other than an accident, which affects or could affect the safety of operations." Accidents and incidents are the error experiences from which airlines have the potential to learn.

EFFECTS OF PRIOR EXPERIENCE ON LEARNING

In the literature on organizational learning there is a large body of work on the learning curve. The learning curve is an empirical finding showing that, in general, experience produces improvement. Early empirical work on the learning curve showed that the log of unit costs tends to decrease linearly with the log of cumulative production volume. So, for example, cumulative production experience tends to lower costs in shipbuilding and automotive production (Argote and Epple, 1990), nuclear power plant production (Zimmerman, 1982), and coal generation (Joskow and Rose, 1985). More recent work has moved away from a focus on cost reduction and productivity improvement to other outcomes of learning. These studies have shown that experience improves customer service and product quality (Dart, Argote, and Epple, 1995; Lapre, Mukherjee, and Van Wassenhove, 2000) and increases the survival rates of hotels (Ingram and Baum, 1997; Baum and Ingrain, 1998) and banks (Kim and Miner, 2000).

In the context of airlines and their errors, it may be that airlines learn from error experience and are able to improve performance over time, reducing subsequent errors (i.e., accidents and incidents). If we look at the airline industry over long time periods, this seems to be the case. Figure 1 plots the accident rate (accidents per 100,000 hours flown) for all U.S. airlines from 1955 to 1997 and exhibits a characteristic learning curve, i.e., as experience accumulates with the passage of time, the error rate declines.

[FIGURE 1 OMITTED]

When individual airlines' accident rates are broken out, as they are in table 1 for some of the larger U.S. airlines, we see the same general decrease in accidents over time as in figure 1, but there is also a fair amount of variance across airlines. For example, from 1957 to 1986, American Airlines had an average of 10.3 accidents per million departures and US Air had 6.6. Variation in airline error rates could come from many sources. One obvious source is the characteristics of the individual airline, e.g., whether it is large or small, the age of its fleet, characteristics of its corporate culture, its management team, and its training procedures. Another possible source of variation, however, is differences in the characteristics of the accidents and incidents experienced by these different airlines. Because experience affects organizational learning, different types of experiences are likely to produce variation in learning rates. One source of differences in experience is whether that experience has homogeneous or heterogeneous causes.

Learning from complexity: effects of prior accidents and incidents on airlines' learning

For all the scientific pizzazz [involved in airline accident investigations], unraveling the subtle, complex chain of events leading to aviation deaths is proving more elusive than ever.

--"Why more plane-crash probes end in doubt," Wall Street Journal, March 22, 1999

Organizations like airlines try to learn from experience, understanding what went wrong so that it won't go wrong next time. But if, as the quote above suggests, the causes are often left in doubt, such learning is likely to be difficult. Learning is also likely to vary across firms, despite industry regulation that should affect all airlines equally. Investigators of the 2000 Air France Concorde crash discovered that British Airways had recommended changes to the Concorde's water deflector in 1995 but that Air France had not made those changes (Phillips, 2000). As Donoghue (1998: 36) explained, "... any safety initiative has an unequal effect on the carriers and becomes an issue to be promoted or fought ... seeking the path that best suits [the airline] individually." Other heavily regulated industries, such as nuclear power, also show substantial variance in incident rates among firms (Morris and Engelken, 1973), which indicates that firms vary in how effectively they learn from their experience. Despite much discussion and analysis of aviation errors (airline accidents and incidents), there has been little work investigating the role of organizational learning and none examining variation in learning across firms in the industry.

Advertisement

Learning from experience has been shown to have important effects on such varied outcomes as manufacturing plant productivity (e.g., Argote, Beckman, and Epple, 1990), service timeliness (Argote and Darr, 2000), and hotel survival (Baum and Ingram, 1998). If firms learn from experience, then the attributes of this experience are likely to affect the rate and effectiveness of learning. Some firms have heterogeneous experience in that their accidents and incidents ("errors") are caused by a large number of different factors, which are likely to interact in complex ways. Some firms have more homogeneous experience, with errors caused by a small number of similar factors. It is likely that the complexity of prior experiences, as well as characteristics of the firms themselves, affect how well airlines can learn from that experience. We investigate these issues in the context of airline accidents and incidents to explain variation in learning among firms in the airline industry.

According to the NTSB (2001) Code of Federal Regulations (49CFR830.2, p. 1195), an accident "means an occurrence ... in which any person suffers death or serious injury, or in which the aircraft received substantial damage." An incident is "an occurrence other than an accident, which affects or could affect the safety of operations." Accidents and incidents are the error experiences from which airlines have the potential to learn.

EFFECTS OF PRIOR EXPERIENCE ON LEARNING

In the literature on organizational learning there is a large body of work on the learning curve. The learning curve is an empirical finding showing that, in general, experience produces improvement. Early empirical work on the learning curve showed that the log of unit costs tends to decrease linearly with the log of cumulative production volume. So, for example, cumulative production experience tends to lower costs in shipbuilding and automotive production (Argote and Epple, 1990), nuclear power plant production (Zimmerman, 1982), and coal generation (Joskow and Rose, 1985). More recent work has moved away from a focus on cost reduction and productivity improvement to other outcomes of learning. These studies have shown that experience improves customer service and product quality (Dart, Argote, and Epple, 1995; Lapre, Mukherjee, and Van Wassenhove, 2000) and increases the survival rates of hotels (Ingram and Baum, 1997; Baum and Ingrain, 1998) and banks (Kim and Miner, 2000).

In the context of airlines and their errors, it may be that airlines learn from error experience and are able to improve performance over time, reducing subsequent errors (i.e., accidents and incidents). If we look at the airline industry over long time periods, this seems to be the case. Figure 1 plots the accident rate (accidents per 100,000 hours flown) for all U.S. airlines from 1955 to 1997 and exhibits a characteristic learning curve, i.e., as experience accumulates with the passage of time, the error rate declines.

[FIGURE 1 OMITTED]

When individual airlines' accident rates are broken out, as they are in table 1 for some of the larger U.S. airlines, we see the same general decrease in accidents over time as in figure 1, but there is also a fair amount of variance across airlines. For example, from 1957 to 1986, American Airlines had an average of 10.3 accidents per million departures and US Air had 6.6. Variation in airline error rates could come from many sources. One obvious source is the characteristics of the individual airline, e.g., whether it is large or small, the age of its fleet, characteristics of its corporate culture, its management team, and its training procedures. Another possible source of variation, however, is differences in the characteristics of the accidents and incidents experienced by these different airlines. Because experience affects organizational learning, different types of experiences are likely to produce variation in learning rates. One source of differences in experience is whether that experience has homogeneous or heterogeneous causes.