Project failure and narratives’ negative emotional content

Organizational performance is often dependent upon the organization engaging in entrepreneurial action (Covin and Miles Reference Covin and Miles1999; Rauch et al. Reference Rauch, Wiklund, Lumpkin and Frese2009). In larger firms, these actions largely begin from R&D activities (Khandawalla Reference Khandawalla1977). It is customary for individuals involved in such projects to develop a sense of psychological ownership toward these projects (Pierce, Kostova, and Dirks Reference Pierce, Kostova and Dirks2001). This feeling of ownership can be seen as a positive psychological resource (Avey, Avolio, Crossley, and Luthans Reference Avey, Avolio, Crossley and Luthans2009), which can in turn result in increases in positive emotions as well as benefits for organizational change (Avey, Wernsing, and Luthans Reference Avey, Wernsing and Luthans2008). Furthermore, positive emotions experienced at the group level have been shown to enhance cooperation, reduce conflict, and positively influence perceived performance (Barsade Reference Barsade2002).

However, these benefits of entrepreneurial projects do not come without potential risks. While entrepreneurial projects can amplify individuals’ motivation for success, they can also result in negative emotional consequences if the projects fail (which, given the high uncertainty surrounding these projects, is a likely outcome) (Shepherd, Covin, and Kuratko Reference 271Shepherd, Covin and Kuratko2009a; Wiklund and Shepherd Reference Wiklund and Shepherd2011). It is possible that if such negative emotions are felt acutely enough at the individual level, they can in turn produce similar emotions within the group as well, essentially forming “collective emotions” (Cannon and Edmondson Reference 265Cannon and Edmondson2001). Collective emotions are emotions that are experienced by a group of individuals within a shared society (Stephan and Stephan Reference Stephan, Stephan and Oskamp2000). Just as individual emotions can influence the group, collectively held emotions can also serve to influence the emotions of the individuals within the group. These group-based emotions “can play a pivotal role in shaping individual and societal responses to conflicting events and in contributing to the evolvement of a social context that maintains the emotional climate and collective emotional orientation that have developed” (Bar-Tal, Halperin, and De Rivera Reference Bar-Tal, Halperin and De Rivera2007). Even though individuals within the group might experience varying personal emotions, it is possible that these commonly held collective emotions can lead group members to common actions that might not necessarily align with their individual emotions but are nonetheless in line with the group perspective.

As they relate to organizational narratives, collective emotions are often evident in the public dialogues organizations develop (Bar-Tal Reference Bar-Tal2001). Excerpts from the annual reports we analyzed (Wolfe and Shepherd, Reference Wolfe and Shepherd2015a) provide evidence of the generation of negative emotions resulting from project failure. An example of this can be seen from an annual report describing the organization's outlook on a failed drug application to the European Medicines Agency (EMA):

This acknowledgment of emotions is important because, as we detail in the following text, these collective negative emotions can have substantial effects on work outcomes and performance (George and Brief Reference George and Brief1992).

As a result of the collective negative emotions generated from project failure, organizations often alter the content of the narratives they construct about failure events (Dutton, Worline, Frost, and Lilius Reference Dutton, Worline, Frost and Lilius2006). Over time, as these collective emotions become more common, they can potentially evolve into organizational routines and norms (Huy Reference Huy1999) as they become ever more ingrained in the organization's systematic response to addressing negative emotions. Furthermore, it is possible that these routines and norms can impact the development of the organizational narrative and result in narratives regarding failure events containing greater levels of negative emotional content. Moreover, these failure narratives could be employed by organizations as a means of justifying or legitimating the activities that produced failed results (Vaara Reference Vaara2002). An example of the use of failure narratives to justify and lend credibility to negative events can be seen in the following passage:

While narratives consist of recounting previous events that are given meaning by the narrative plot, we note that the plot is not intrinsic to the events but rather is a construct used by the authors in an attempt to make sense of these occurrences (Vaara Reference Vaara2002). As a result, there is the potential to alter a particular narrative of project failure in the hopes of painting the experience in a more favorable, less disappointing manner.

More negative emotional content of a narrative about project failure could signal to key shareholders that even though these project efforts resulted in failure, the projects were important and that the firm remained optimistic about eventual future success regarding these projects. In addition, these messages about the importance of failure could communicate continued organizational commitment to R&D activities, which, as previously stated, are the key to future performance of many firms in knowledge-intensive industries. The following is an excerpt showing how an organization utilizes a narrative regarding a failure event as a mechanism to communicate future commitment to R&D activity as well as optimism for the eventual success of the failed research initiatives:

Even though we have begun to explore the importance of negative emotional content in narratives (here and earlier in the book), such a discussion is incomplete without also exploring the role of positive emotions, to which we now turn.

Narratives’ positive and negative emotional content

Although positive and negative emotions are often regarded as related concepts, they remain distinctly separate and unique constructs. Substantial research has been done in this area and shows strong evidence that positive and negative emotions are largely independent of each other (Diener and Emmons Reference 266Diener and Emmons1984). Positive and negative emotions can be examined more closely based on their two underlying dimensions – frequency and intensity – and when this is done, the distinction between the two becomes clearer (Diener, Larsen, Levine, and Emmons Reference Diener, Larsen, Levine and Emmons1985). Of course, time spent experiencing one emotion impacts the amount of time individuals can experience other emotions, and because of this, research has shown that in regard to the frequency dimension, the two forms of emotions are highly negatively correlated (Diener et al. Reference Diener, Larsen, Levine and Emmons1985). Conversely, a predilection to experience one type of emotion more intensely can result in the tendency to experience another type of emotion more intensely as well. This results in the intensity dimension of both types of emotions being highly positively correlated (Diener et al. Reference Diener, Larsen, Levine and Emmons1985). Therefore, because the average levels of affect one experiences are a result of both frequency (for which positive and negative emotions are negatively correlated) and intensity (for which positive and negative emotions are positively correlated), the overall correlations of positive and negative emotions are low. The independent nature of these two emotional constructs allows us to explore when and how they are related and what differential effects they have.

Positive emotions appear to be able to decrease the consequences of negative emotions via an “undoing” mechanism (Fredrickson, Mancuso, Branigan, and Tugade Reference Fredrickson, Mancuso, Branigan and Tugade2000). The undoing role of positive emotions refers to the ability that positive emotions have to reduce and/or eliminate the detrimental consequences experienced as a result of negative emotions (Fredrickson et al. Reference Fredrickson, Mancuso, Branigan and Tugade2000). As evidence of this undoing effect of positive emotions, Fredrickson and colleagues conducted a study of individuals’ resilience following the terrorist attacks on 9/11 and found that positive emotions substantially benefited individuals by enhancing their overall resilience following negative experiences (Fredrickson, Tugade, Waugh, and Larkin Reference Fredrickson, Tugade, Waugh and Larkin2003).

Over and above undoing negative emotions, it appears that positive emotions can “broaden one's thought–action repertoire, expanding the range of cognitions and behaviors that come to mind” (Tugade and Fredrickson Reference Tugade and Fredrickson2004). By undoing negative emotions and broadening thought–action resources, positive emotions might prove to be helpful in regulating the negative emotions that can develop as a result of failure (Tugade and Fredrickson Reference Tugade and Fredrickson2004). For example, positive emotions have been shown to increase individual resilience (Tugade and Fredrickson Reference Tugade and Fredrickson2004), enhance the likelihood that failed entrepreneurs will put forth the effort to attempt new ventures (Hayward, Forster, Sarasvathy, and Fredrickson Reference Hayward, Forster, Sarasvathy and Fredrickson2009), and assist organizations in accomplishing organizational transformation (Sekerka and Fredrickson Reference Sekerka, Fredrickson, Ashkanasy and Cooper2008).

Some passages demonstrating positive emotional content from our sample include the following statement regarding an organization's hopeful perspective in terms of its developmental pipeline:

This passage regarding one firm's positive outlook about their projected future growth also demonstrates this positive emotional content:

Positive emotions and their effects (i.e., undoing and broadening) have been shown to exist at the organizational level (Fredrickson Reference Fredrickson, Cameron, Dutton and Quinn2003). At the organizational level, positive emotions can stimulate “useful cognitive and social capabilities” (Sekerka and Fredrickson Reference Sekerka, Fredrickson, Ashkanasy and Cooper2008), enhance cooperation as a result of building and reinforcing relational strength (Sekerka and Fredrickson Reference Sekerka, Fredrickson, Ashkanasy and Cooper2008), and amplify the potential for organizational change (Avey et al. Reference Avey, Wernsing and Luthans2008). Specifically, emotions evolve in social settings, and as such, they can perform social functions (Fischer and Manstead Reference Fischer, Manstead, Lewis, Haviland-Jones and Barrett2008). For example, it has been suggested that positive emotions can serve as a form of “social glue,” enhancing group members’ affiliation with each other (Fischer and Manstead Reference Fischer, Manstead, Lewis, Haviland-Jones and Barrett2008), and studies have shown that positive emotional expressions by leaders or managers are predictive of overall group performance (George Reference George1995). Nevertheless, how do these emotions and their effects play out in the use of organizational narratives for sensemaking purposes? We now turn to addressing this question.

Positive feedback and narratives’ negative emotions

Although project failure is often perceived as a negative event (Iacovoc and Dexter Reference Iacovoc and Dexter2005; Shepherd et al. Reference Shepherd, Patzelt and Wolfe2011b), the failure of one or more projects does not necessarily reflect the overall performance of the firm. Indeed, it has been argued that performance is enhanced when organizations are more entrepreneurial, which involves experimenting more, probing the future, and quickly terminating projects that do not show promise (McGrath Reference McGrath1999). In essence, a higher frequency of project failure is common in firms that engage in greater levels of proactiveness and innovation (McGrath Reference McGrath1999), both of which comprise the fundamental building blocks of an overall entrepreneurial orientation (Covin, Green, and Slevin Reference Covin, Green and Slevin2006). Furthermore, and as reinforced consistently throughout this book, failures represent an opportunity for individual and organizational learning (Chuang and Baum Reference Chuang and Baum2003), which are important antecedents to organizational success (Minniti and Bygrave Reference Minniti and Bygrave2001). Learning as a result of failure affords firms the opportunity to make sense and understand the underlying causes of those failures, thereby motivating future action and hopefully reducing the probability that similar mistakes will be repeated.

Because profitability has been shown to be such an important indicator of firm performance (Anderson, Fornell, and Lehmann Reference Anderson, Fornell and Lehmann1994), increases in profitability are typically considered to be a positive event – namely, positive feedback indicative of improved firm performance. Positive feedback can produce a number of advantages, including increased group pride, involvement, and esteem (Nadler Reference Nadler1979), which can in turn facilitate member cooperation and collaboration (Druskat and Wolff Reference Druskat, Wolff, Cherniss and Goleman2001), reduce negative emotions, and enhance positive well-being (Salanova, Llorens, Cifre, Martínez, and Schaufeli Reference Salanova, Llorens, Cifre, Martínez and Schaufeli2003). As a result, in organizations where failure is likely to occur, it will be important for the organization to focus its attention on other sources of positive feedback, such as firm profitability, to counteract the potentially negative consequences that high project failure rates can have at both the individual and organizational levels.

Most often, organizations determine whether or not they are performing at an acceptable level by comparing their current performance with previously established thresholds (Greve Reference Greve2002). The results of this evaluation often impact important organizational processes, such as innovation (Bolton Reference Bolton1993), strategic perspective (Grinyer and McKiernan Reference Grinyer and McKiernan1990), and corporate re-orientation (Lant, Milliken, and Batra Reference Lant, Milliken and Batra1992). When analyzing performance from a threshold standpoint, organizations typically adopt the view that if current performance is above that of prior years, it is deemed acceptable and successful. However, if performance dips below historical levels, performance is typically viewed in a negative light (Greve Reference Greve2002). Therefore, as long as the firm's profitability remains at or above a threshold level, the risks of the negative consequences associated with high failure rates are somewhat mitigated by this positive feedback. Indeed, positive performance feedback (despite a high failure rate) has been found to encourage more entrepreneurial endeavors, such as investments in R&D (see also Scherer Reference Scherer2001; Wolfe and Shepherd Reference Wolfe and Shepherd2015k).

It is likely that the narratives organizations construct will be modified to include information from performance feedback and to portray firms’ overall position. To investigate this issue, we (Wolfe and Shepherd, Reference Wolfe and Shepherd2015k) analyzed annual reports from public firms that had submitted at least one drug-approval application to the EMA from 2002 to 2010. Because the performance events (e.g., project failure and profitability) occurred before narrative development (i.e., the construction of the annual report), there is a theoretical basis behind the causal relationship between these variables. Analyzing annual reports also proved advantageous because they represent narratives that are closely examined by important stakeholders and can therefore have important consequences for organizations (Stanton, Stanton, and Pires Reference Stanton, Stanton and Pires2004). As a result, it is vital to top management that the narratives created within these reports appropriately portray the organization at hand as well as justify its past, present, and future activities. Our sample included 68 firms and a total of 356 observations (i.e., total number of annual reports).

In our study of how performance events – both negative (in the form of project failure) and positive (in the form of increases in profitability) – related to levels of positive and negative emotional content within organizational narratives, we found that both positive and negative performance events influenced the negative content of organizational narratives. Additionally, we discovered that not only did positive emotional content influence narratives’ negative emotional content but that this relationship was moderated by gains or losses in profitability.

In this study, we found that even for large organizations, which usually construct relatively sanitized and standardized (e.g., mentions of failures restricted, negative emotional content minimized, etc.) narratives, such as annual reports, increased levels of project failure resulted in higher levels of negative emotional content. Additionally, narratives with more positive emotional content in general contained less negative emotional content. Moreover, Wolfe and Shepherd (Reference Wolfe and Shepherd2015a) found evidence that the negative relationship between the positive emotional content and negative emotional content of failure narratives was greater when there was high positive performance feedback than when there was low positive performance feedback. These findings have implications for our understanding of narratives’ emotional outcomes related to performance feedback.

Project failure and narratives’ negative emotional content

The first important implication of our findings is that the greater the failure rate of entrepreneurial projects, the greater the narrative's negative emotional content (controlling for narrative length) (Wolfe and Shepherd, Reference Wolfe and Shepherd2015a). Although the development of new products and the R&D process in general are the foundation for success in the pharmaceutical industry (Del Monte and Papagni Reference Del Monte and Papagni2003; DiMasi et al. Reference DiMasi, Hansen and Grabowski2003), these projects by their very nature are higher in uncertainty and therefore result in higher project failure rates (DiMasi Reference DiMasi2001). This is important to note because it has been shown that employees can develop high levels of psychological ownership of projects under their control (Pierce et al. Reference Pierce, Kostova and Dirks2001), which can in turn generate positive emotions that facilitate individual and organizational performance, particularly in situations of organizational change (Avey et al. Reference Avey, Wernsing and Luthans2008). However, as established in Chapter 2, the more important these projects are to project team members, the more grief they are likely to experience when the projects fail (Shepherd et al. Reference 271Shepherd, Covin and Kuratko2009a; Shepherd, Patzelt, and Wolfe Reference Shepherd, Patzelt and Wolfe2011a).

This type of negative emotional reaction can be felt at both the individual and the collective levels (Cannon and Edmondson Reference 265Cannon and Edmondson2001). As mentioned earlier, collective emotion refers to “emotions that are shared by a large number of individuals in a certain society” (Wolfe and Shepherd Reference Wolfe and Shepherd2015k). The development of collective emotions from individual emotions can occur both implicitly and explicitly via a number of emotional processes. Implicitly, collective emotions can develop through emotional contagion (Hatfield et al. Reference Hatfield, Cacioppo, Rapson and Clark1992) and vicarious affect (Bandura Reference 264Bandura1986). In terms of explicit processes, both affective influence (Ashkanasy and Tse Reference Ashkanasy, Tse, Ashkanasy, Hartel and Zerbe2000) and affective emotional displays (Kelly and Barsade Reference Kelly and Barsade2001) can contribute to the development of collective emotions. Regardless of the mechanisms by which individual emotions become collective emotions, collective emotions arising from project failure can alter a narrative (Dutton et al. Reference Dutton, Worline, Frost and Lilius2006).

Not surprisingly, because project failures often generate negative emotions (Cannon and Edmondson Reference 265Cannon and Edmondson2001; Shepherd et al. Reference Shepherd, Patzelt and Wolfe2011a), the narratives constructed about those failures will have high negative emotional content. However, increased negative emotional content might not be the only way project failure influences organizational narratives. Because a narrative plot is constructed by an author in an attempt to make sense of past events (Vaara Reference Vaara2002), it is possible that the organization might use a project failure narrative to explain the activities leading up to the failure (Vaara Reference Vaara2002), modifying the plot as needed to alter how project failure is portrayed. That is, rather than simply reflect the consequences of the failure, the narrative might construct and reflect the antecedents to the project failure. In this manner, narratives could be used to signal to stakeholders the importance of failed projects to the firm and/or the firm's continued interest in similar entrepreneurial projects as a means to enhance firm performance.

An example of a passage containing higher levels of negative emotions as a result of project failure can be seen in the following passage detailing one organization's disappointment in having to withdraw a product from the market as well as ceasing development of another late-stage compound:

Positive and negative emotional narrative content

A second important implication of our findings is that the greater the positive emotional content of the narrative, the lower the negative emotional content (again controlling for narrative length) (Wolfe and Shepherd, Reference Wolfe and Shepherd2015a). Such a finding is consistent with the notion that positive emotions can “undo” the effects of negative emotions (Fredrickson et al. Reference Fredrickson, Mancuso, Branigan and Tugade2000) as well as broaden the set of thought–action resources, which may help regulate negative emotions generated by an unexpected failure (Tugade and Fredrickson Reference Tugade and Fredrickson2004). In addition to the effects that positive emotions can have at the individual level, evidence suggests that they can prove to be similarly beneficial at the organizational level (Fredrickson et al. Reference Fredrickson, Cameron, Dutton and Quinn2003). Emotions develop in social contexts and can thus influence social functions (Fischer and Manstead Reference Fischer, Manstead, Lewis, Haviland-Jones and Barrett2008). In this sense, positive emotions have been proposed to operate as a type of “social glue” that binds group members together on a social level (Fischer and Manstead Reference Fischer, Manstead, Lewis, Haviland-Jones and Barrett2008) and thereby increases the overall level of cooperation exhibited within the organization (Sekerka and Fredrickson Reference Sekerka, Fredrickson, Ashkanasy and Cooper2008). This increased cooperation has been shown to be particularly useful in situations involving organizational change (Avey et al. Reference Avey, Wernsing and Luthans2008).

Examples of passages containing positive emotional elements include the following, wherein an organization is discussing positive results related to its drug-development activities:

Additionally, excerpts demonstrating the use of positive and negative emotional content within the same context can be seen in the following passage detailing disappointment at results of a drug trial in the United States but continued optimism for the drug's applications in East Asian markets:

Moderating role of profitability

Finally, the arrangement of narratives’ positive and negative emotional content can be complex. Indeed, when positive performance events, such as increases in profitability, do occur, the relationship between positive and negative narrative content is amplified (Wolfe and Shepherd, Reference Wolfe and Shepherd2015a). Positive outcomes in areas other than NPD might serve to mitigate the likelihood that individuals and organizations will over-identify with and ruminate over project failures within the organizational narrative. Such positive events can shift attentional focus away from the negative consequences of failure and onto other areas of importance to the organization (Ocasio Reference Ocasio1997, Reference Ocasio2011). This broadening of attention might reduce or even eliminate the possibility that project failures will become the driving focus of attention, thereby decreasing the potential for organizations to include high levels of negative emotional content in the narratives they construct. From an entrepreneurial point of view, this arrangement of emotional content – focusing more on the positive rather than the negative – provides support for the notion that “entrepreneurial action is characterized by the dominance of pride and the relative absence of shame” (Goss, Reference Goss2005: 212). To better visualize the moderating role that profitability has on the relationship between a narrative's positive and negative emotional content, Figure 8.2 contains a graphical representation of this effect. As is evident in Figure 8.2, the negative influence that positive emotional content has on negative emotional content is more pronounced when organizations have experienced gains in profitability rather than losses in profitability.

Figure 8.2: Positive emotional content, profitability, and negative emotional content

Entrepreneurial orientation content and bouncing back

Although the majority of previous evidence has supported the idea that EO has a positive effect on organizational performance (Rauch et al. Reference Rauch, Wiklund, Lumpkin and Frese2009), recent research has suggested that this relationship is perhaps not linear (Zahra and Garvis Reference Zahra and Garvis2000). In fact, a recent project examining the relationship between EO and performance in small- to medium-sized firms indicated that this relationship is U-shaped in nature (Kreiser, Marino, Kuratko, and Weaver Reference Kreiser, Marino, Kuratko and Weaver2013). Adopting a sensemaking perspective, narratives can be used to communicate the necessary amount of EO needed to effectively bounce back after failure. Because narratives give meaning to events (Orbuch Reference Orbuch1997), they can be utilized not only to relate details about past events but also to influence future behaviors (Orr Reference Orr1995). In this manner, they are particularly salient influences driving organizational change (Dunford and Jones Reference Dunford and Jones2000). Thus, the level of EO content contained within a narrative signifies an attempt to communicate how failure is interpreted and can thus have an important influence on subsequent performance.

An important point to make note of is that undertaking EO activities does not universally produce positive results (Hart Reference Hart1992). Because entrepreneurial initiatives usually require considerable commitment of initial resources (Hornsby, Kuratko, Shepherd, and Bott Reference 268Hornsby, Kuratko, Shepherd and Bott2009), it is possible that organizations might forego investing in activities critical for organizational survival in order to engage in entrepreneurial activities (Rosenbusch, Brinckmann, and Bausch Reference Rosenbusch, Brinckmann and Bausch2011). Therefore, in terms of attempting to survive, organizations that have a relatively low EO may be better able to allocate resources to other more beneficial areas and may consequently experience increased performance. This low EO would be reflected in narratives containing fewer examples of language relating to the need to act entrepreneurially (e.g., proactive, innovative, risk taking, etc.). Examples of such language indicating a low need to act entrepreneurially include this from Coach Lynch after Indiana University's first loss of the season to Southern Illinois University:

Similarly, Coach Roof of Duke University noted the following after his team's first loss of the Reference Roof2006 football season:

Teams that act proactively (Lumpkin and Dess Reference Lumpkin and Dess2001) or engage in riskier behaviors in an attempt to increase performance (Covin and Slevin Reference Covin and Slevin1989) often find that such actions come with considerable costs (Li and Atuahene-Gima Reference Li and Atuahene-Gima2001; Rosenbusch et al. Reference Rosenbusch, Brinckmann and Bausch2011). Because EO-related initiatives usually necessitate large initial expenditures of key resources (Hornsby et al. Reference 268Hornsby, Kuratko, Shepherd and Bott2009) and are frequently linked with higher levels of uncertainty (Alvarez Reference Alvarez2007), these types of entrepreneurial activities have a lower chance of producing rewards that outweigh the costs for the organization (Freel Reference 267Freel2005). Indeed, moderate levels of EO narrative content could signal a lack of commitment to innovate and change, which could in turn translate into a lack of success in subsequent performances. By only moderately varying their activities, teams might suffer from a reduced ability to execute these activities because they have not had sufficient time or motivation to develop a deep understanding of the requisite changes. As a result, it is possible that the variations introduced through EO initiatives might not be different enough to confuse competitors, which could hinder performance.

While moderate levels of EO might lead to the costs outweighing the benefits, it is possible that teams willing to make substantial commitments to EO activities will experience benefits that outshine the associated costs (Kreiser et al. Reference Kreiser, Marino, Kuratko and Weaver2013). For teams that are highly committed to entrepreneurial initiatives, the costs are still substantially high. However, because of such teams’ likely high level of commitment to and focus on their intended goals, the benefits of these actions could ultimately be greater than the initial costs. Examples of high EO content found within our sample include excerpts like the following from Coach Alvarez of Wisconsin following his team's loss to Northwestern in Reference Alvarez2005:

Another example of higher EO content can be seen in the following passage from Coach Spurrier of South Carolina:

In summary, while the extant literature has stressed that EO is beneficial in general, recent investigations suggesting that this relationship is not linear in nature have lead us to further examine how EO influences a team's ability to bounce back from previous failure, to which we now turn. Our results indicate that in fact the relationship between narrative EO content and subsequent performance is U-shaped, with relatively low or high levels of EO content proving to be more beneficial than moderate levels of EO content. For a clearer representation of this relationship, see Figure 8.3.

Figure 8.3: Entrepreneurial orientation and subsequent performance

Negative emotions and bouncing back

Throughout previous chapters, we have established that entrepreneurial activities are inherently risky endeavors (Covin and Slevin Reference Covin and Slevin1991; Lumpkin and Dess Reference Lumpkin and Dess1996) and are thus susceptible to failure (Wiklund and Shepherd Reference Wiklund and Shepherd2011). We have also shown that failures of this nature can generate substantial negative emotional reactions among those involved (Shepherd et al. Reference Shepherd, Patzelt and Wolfe2011b). These negative emotional reactions to project failure can have a positive impact on subsequent performance. That is, negative emotions are not always detrimental. Evidence has shown that negative emotions can help maintain attentional focus (Bless, Bohner, Schwarz, and Strack Reference Bless, Bohner, Schwarz and Strack1990), enhance information processing (Schwarz Reference Schwarz, Higgins and Sorrentino1990), improve spatial performance (Gray Reference Gray2004), increase perceived acceptable performance standards (Cervone, Kopp, Schaumann, and Scott Reference Cervone, Kopp, Schaumann and Scott1994), and amplify the likelihood of eventual goal attainment (Brown and Eisenhardt Reference Brown and Eisenhardt1997). Based on this evidence, it is likely that a complete absence of (or very low levels of) negative emotional content in narratives relating to failure might not be beneficial and could actually hamper subsequent performance.

While expressing little to no negative emotions regarding failure can prove detrimental to subsequent performance, excessive levels of negative emotions can also lead to negative consequences at both the individual (Pierce et al. Reference Pierce, Kostova and Dirks2001) and group levels (Cannon and Edmondson Reference 265Cannon and Edmondson2001). Evidence has indicated that over-identification with or rumination over failure can interfere with problem solving (Lyubomirsky and Nolen-Hoeksema Reference Lyubomirsky and Nolen-Hoeksema1995) and impede attempts to understand and make sense of the failure event (Borders, Earleywine, and Jajodia Reference Borders, Earleywine and Jajodia2010; Kross, Ayduk, and Mischel Reference Kross, Ayduk and Mischel2005). For these reasons, while a move from experiencing and expressing low levels of negative emotions to more moderate levels might prove beneficial, moving from moderate to high levels of negative emotional content in a failure narrative will likely result in decreased performance. Therefore, a moderate amount of negative emotional content within a failure narrative serves to adequately communicate information regarding future actions without focusing too much attention on the failure (i.e., considerable attention allocated to failure can lead to rumination and reduced performance) (Lyubomirsky and Nolen-Hoeksema Reference Lyubomirsky and Nolen-Hoeksema1995).

Interestingly, contrary to these expectations, the results of our recent study (Wolfe and Shepherd Reference Wolfe and Shepherd2015k) do not substantiate the idea that moderate levels of negative emotional narrative content are the most beneficial for subsequent performance. In fact, the results of our recent study indicate that either high or low levels of negative emotional content produce better performance outcomes than moderate levels. It appears that low negative emotions can help maintain affective commitment to the organization (Belschak and Hartog Reference Belschak and Hartog2009) – affective commitment to the team in this case – and affective commitment has been associated with enhanced performance outcomes (Sinclair, Tucker, Cullen, and Wright Reference Sinclair, Tucker, Cullen and Wright2005). Examples of low-level negative emotional statements include passages like “You know, we didn't get the result that we hoped for today … but I want our players not to hang their heads”; “We are not satisfied obviously with losing the game”; and “We've got to work a little bit harder and come together a little bit more… . It's only devastating if you let it be devastating.” While negative emotions have been associated with reduced affective commitment (Belschak and Hartog Reference Belschak and Hartog2009; Shepherd et al. Reference Shepherd, Patzelt and Wolfe2011b) and inhibited problem-solving ability (Lyubomirsky and Nolen-Hoeksema Reference Lyubomirsky and Nolen-Hoeksema1995), both of which are believed to diminish performance (Belschak and Hartog Reference Belschak and Hartog2009; Rude, Maestas, and Neff Reference Rude, Maestas and Neff2007), it appears that – at least in the short run – an obsession with failure (reflected in ruminations Rude et al. Reference Rude, Maestas and Neff2007) may serve as a strong motivator to bounce back and perform well in the next project – namely, next week's game.

Excerpts from our study that exhibit higher levels of negative emotions include the following quote from Coach Smith of Michigan State University following their first loss of the 2005 season:

This excerpt from Coach Mullen of Mississippi State University following their first loss in Reference Mullen2009 also illustrates this point:

Our findings indicate that indeed relatively low or high levels of negative emotional content contained within the failure narrative were more beneficial to subsequent performance than were moderate levels of negative emotional content. A graph of this U-shaped relationship between the failure narrative's negative emotional content and subsequent performance can be seen in Figure 8.4.

Figure 8.4: Negative emotional content and subsequent performance

Positive emotions and bouncing back

Just as it is important to understand the role negative emotions play in the relationship between failure narratives and subsequent performance, positive emotions’ influence must also be taken into consideration. As touched upon earlier in this chapter, positive emotions can result in a “broaden-and-build” phenomenon (Fredrickson Reference Fredrickson1998, Reference Fredrickson2001). They have been shown to enhance individuals’ performance in work-related tasks (Lyubomirsky, King, and Diener Reference Lyubomirsky, King and Diener2005) as well as in entrepreneurial tasks (Cardon, Zietsma, Saparito, Matherne, and Davis Reference Cardon, Zietsma, Saparito, Matherne and Davis2005; Cardon, Wincent, Singh, and Drnovsek Reference Cardon, Wincent, Singh and Drnovsek2009). By facilitating information integration through the broaden-and-build principle, positive emotions have been shown to improve individual reasoning and decision-making processes (Lyubomirsky et al. Reference Lyubomirsky, King and Diener2005). Furthermore, similar effects have been demonstrated to occur at the group level (Fredrickson Reference Fredrickson, Cameron, Dutton and Quinn2003). Therefore, with regard to the positive emotional content of failure narratives, it is possible that including moderate levels of positive emotional content results in improvements to subsequent performance.

While positive emotions appear to enhance performance, their influence might not always be beneficial. The reasoning behind this lies in the relationship between positive emotions and optimism. Evidence suggests that positive emotions can increase the level of optimism after negative events (Fredrickson et al. Reference Fredrickson, Cameron, Dutton and Quinn2003), which can prove useful because positive emotions, such as optimism, have been shown to enhance performance following negative events at both the individual (Isen Reference Isen2002) and organizational (Avey et al. Reference Avey, Wernsing and Luthans2008) levels. However, high levels of optimism (i.e., over-optimism) can lead to poor decision making (Hmieleski and Baron Reference Hmieleski and Baron2008). Indeed, over-optimism is perhaps the most commonly noted bias among entrepreneurs (Baron Reference Baron2004), and its negative influence on entrepreneurial performance is well documented (Lowe and Ziedonis Reference Lowe and Ziedonis2006). The danger in developing over-optimism is that it can result in reduced ability and motivation to seek out and recognize the need for change and improvement. This blindness to the necessity of change and/or lack of motivation to do anything about it are the antithesis of the learning and change necessary to avoid project failure. Over-optimism can make project failure more likely by obstructing or delaying project adaptations. Therefore, although expressing some positive emotions regarding a failure event can produce benefits to subsequent performance, expressing extreme levels of positive emotions in a failure narrative could actually hinder future actions.

Examples of such passages from our study include the following from Coach Callahan of Nebraska following his team's first loss of the Reference Callahan2005 season:

This passage from Coach Stoops of Oklahoma University following his team's first loss of the Reference Stoops2008 season also demonstrates extreme positive emotions:

Our findings support the proposed inverted U-shaped relationship between the failure narrative's positive emotional content and subsequent performance, where moderate levels of positive emotional content are more beneficial than relatively high or low levels with regard to their influence on subsequent performance. A graph of this relationship can be seen in Figure 8.5.

Figure 8.5: Positive emotional content and subsequent performance

Implications for entrepreneurship, failure, and organizational narratives

The findings of our first study represent important contributions to the study of corporate entrepreneurship, particularly as it relates to science-based organizations. It is imperative for science-based organizations to continue to maintain a high level of activity in R&D areas, for which technological feasibility is questionable at best and innovative activities (e.g., R&D) face inordinately high failure rates. In finding that organizations alter their narratives’ content as a result of certain performance events, we presented evidence of how organizations attempt to understand these events and presumably apply that understanding to help maintain entrepreneurial and innovative activities. Additionally, our findings that failure rates alone are not the only important factor to consider with regard to narratives’ levels of negative emotional content, we present evidence that not all failures are created equal and that failure events should not be viewed as homogeneous, universally identical experiences.

In addition, the results of our first study also indicate that if positive performance feedback occurs concurrently with failure, it is possible that organizations might choose to focus on communicating the positive aspects of their prior performance, essentially shifting focus away from project failures. This suggests that the interplay between positive and negative emotional content within organizational narratives is indeed complex in nature. Organizations must weigh a number of important factors when determining the configuration of these elements within the narratives they construct. This is particularly salient as the messages communicated within those organizational narratives will be influential in determining exactly how organizations understand and make sense of past events as well as how they enact and influence future actions.

Implications for failure, sensemaking, and subsequent performance

Our second study extended the findings of the first, examining the role that failure narratives play in sensemaking and how specific elements contained within these narratives influence bouncing back with regard to subsequent performance. With regard to evidence supporting our proposed U-shaped relationship between a failure narrative's EO content and subsequent performance, we believe this represents an important enrichment of our understanding of the EO–performance relationship. Unlike most previous research, which has focused on EO as a relatively stable and enduring characteristic, our study examined how EO relates to subsequent performance over a much shorter time frame. From this perspective, our results suggest that in order to effectively bounce back quickly, it is perhaps more beneficial to have a relatively high or low EO perspective rather than a more moderate EO outlook, essentially allowing for the opportunity to “stay the course” or “right the ship” depending on the situation at hand.

A second important implication with regard to this study can be seen in the finding of a similar U-shaped relationship between negative emotional content and subsequent performance. Whereas most conventional wisdom would point to the opposite relationship holding true, with moderate levels of negative emotions proving more beneficial than extremely high or low levels, our results suggest the exact opposite. Again, this could be related to the relatively short window of time over which these influences are measured. While high levels of negative emotions can lead to obsessive ruminations regarding a failure event, which can prove very detrimental over longer periods of time, such obsession might actually be beneficial in motivating individuals and organizations to bounce back quickly with regard to short-term performance. Therefore, rather than contradicting the existing prevailing thought that high levels of negative emotions adversely influence performance, our results provide a complementary perspective with regard to the potentially differential influence that negative emotions can have on failure over short versus long periods of time.

Finally, our results complement and extend existing literature regarding the influence of positive emotions. While we did find that positive emotions do enhance subsequent performance, this influence only holds true to a certain level, and in fact, further increases past this “optimal level” can actually result in diminished performance. This inverted U-shaped relationship between positive emotions and subsequent performance aligns with the idea that while low levels of positive emotions might restrict the cognitive processes underlying the narrative, essentially lowering the potential to change as a result of failure, high levels of positive emotions can produce heuristic biases (e.g., over-optimism) that can impede performance as well. As a result, this study provides evidence of a potential boundary condition to the broaden-and-build influence of positive emotions, indicating that higher levels of positive emotions might not always prove to be more beneficial – at least when it pertains to failure narratives’ emotional content and their ability to influence bouncing back quickly with regard to subsequent performance.