The Development of Dynamic Capability Over Time: A ...



Learning, Environmental Dynamism

and the Evolution of Dynamic Capabilities

Abstract

This paper presents a simulation model of the development of knowledge, routines and dynamic capability in organizations. We draw on system dynamics to explore trade-offs arising in the decision to invest in deliberate learning processes to enhance the development of dynamic capabilities. Four different effects, namely the tool utility and consciousness (positive) effects as well as the inertia and experience base (negative) effects on dynamic capability are considered. The model also distinguishes the two forms of deliberate learning (knowledge articulation and codification) along these effects as well as cognitive and other resource constraints. The simulation results show differing levels of effectiveness as environmental dynamism increases. Knowledge codification appears to be the optimal strategy at intermediate levels and only a combined articulation and codification approach is capable to maintain the firm’s ability to adapt its operating routines in highly turbulent environments.

Introduction

The study of dynamic capabilities, the organization’s capacity to change its operations and adapt them to the environmental requirements, has taken center stage in the debate on strategic management as well as organization theory (Teece, Pisano and Shuen 1997, Eisenhardt and Martin 2000, Zollo and Winter 2002, Winter 2003, Helfat and Peteraf 2003). The notion, which has received several, and only partially aligned, definitions, lies at the heart of of the organization’s ability to enact change in a systematic and fruitful way. Winter (2003) clarifies that organizational change happens in one of two ways: the first with ad-hoc problem driven search, and the second through the action of “stable patterns of activity aimed at creating or changing operating routines in pursuit of enhanced organizational effectiveness”, the definition of dynamic capabilities in Zollo and Winter (2002).

Whereas the question 'what are dynamic capabilities?' seems to have found increasing convergence in its response (cf. Helfat et al. 2006), the related question 'how do dynamic capabilities develop?' remains open to debate and scholarly inquiry. Zollo and Winter (2002) offer a conceptual model that takes into explicit consideration the role of intentionality in the learning process, distinguishing between semi-automatic learning (e.g. learning-by-doing) and deliberate types of learning (i.e. knowledge articulation and codification). They argue the evolution of both dynamic capabilities and operating routines, through a recursive cycle of variation, selection, replication and retention processes, is fundamentally determined by the relative effectiveness of these learning mechanisms.

The following question, however, remains: what influences the relative effectiveness of these different learning mechanisms in producing this particular type of change capacity? Under what conditions, in other words, are investments in deliberate learning processes warranted?

Whereas Zollo and Winter (2002) focus on task characteristics as moderating factors for the relative effectiveness of deliberate learning investments, this paper considers environmental dynamism as a key condition to explore the role of learning mechanisms in the evolution of dynamic capabilities. The driving question in this study, therefore, is: to what extent and how does the degree of dynamism in environmental change influence the development of the firms’ ability to adapt its operating processes to the new demands and conditions of their environment?

The question is tackled with a simulation model based on system dynamics principles. We adopt system dynamics modeling because assumptions are (a) made explicit and (b) observed simultaneously in their interdependent and non-linear influences on the outcome under study (e.g. Rudolph and Repenning 2002, Sterman 2000). Moreover, system dynamics modeling is built on the explicit distinction and analysis of (both tangible and intangible) stocks and flows, which are acknowledged to be increasingly important in developing our understanding of organizational knowledge and dynamic capabilities (Helfat 1997, Helfat and Peteraf 2003). System dynamics suggests that any kind of capability can be modeled as a stock, or a set of related stocks, that accumulates or depletes over time as a result of flows in and out of this stock (Sterman 2000). As such, the model developed in this paper attempts to bridge and integrate the stock versus flow conceptualization of dynamic capability.

In order to focus on the development of this specific kind of capabilities in a parsimonious and tractable way, the model presented below will make two important simplifying assumptions. First, that the development of dynamic capabilities is first and foremost an internal process, that it is not influenced in a significant way by external or institutional pressures, other than an aggregate variable 'environmental dynamism'. Second, we assume the effectiveness of the learning mechanisms underlying the evolution of dynamic capabilities are substantially independent from ad-hoc problem driven processes that constitute the most diffused and frequent way organizations cope with environmental dynamism.

In the next section, we will develop the theoretical arguments underlying the model. Subsequently, key features of the simulation model are described. We then run several simulation experiments with the model and conclude with interpretations for theory development, empirical inquiry and management practice.

Theoretical Background

Previous studies have focused on dynamic capability as arising from routines for variation, selection, replication and retention (e.g. Helfat and Raubitschek 2000, Zollo and Winter 2002, Zott 2003). For example, the simulation model developed by Zott (2003) shows how certain attributes of dynamic capability contribute to the emergence of differential firm performance within the same industry. We are particularly interested in the intra-organizational tradeoffs inherent in the evolution of dynamic capability. In this respect, Zollo and Winter (2002) distinguish between three types of knowledge (processes) that interact to produce dynamic capability:

• Tacit experience that represents the accumulation of lessons learnt directly from external stimuli.

• Articulated knowledge which consists of the result of efforts at articulating experiential knowledge.

• Codified knowledge which, in a similar vein, is built up of the codification of articulated knowledge.

For example, a consulting firm has an experiential knowledge base that consists of the sum of all (partly tacit) experience embedded in its consultants. In addition, it has an articulated knowledge base residing in the heads of the individual consultants that took form through contact with others, taking in others’ representations and by articulation efforts of one’s own (e.g. by brown bag seminars, project reviews, and senior-junior mentoring). Finally, a subset of articulated knowledge is being codified (e.g. in formal audits, manuals, procedures, tools) on some kind of independent medium like paper, hard disks, and information databases.

In this respect, Hansen, Nohria and Tierney (1999) observe that consulting firms can employ different knowledge management strategies. In many consulting firms knowledge is closely tied to the person who developed it and is shared mainly through direct person-to-person communication; knowledge codification plays a minor role in these firms. In other firms more attention is given to codifying and storing knowledge in databases, where it can be easily accessed and used by anyone in the company. Hansen et al. (1999) suggest that, to be effective in the consulting business, a firm needs to excel in one of the two strategies and use the other strategy in a supporting role. A similar need to focus on one particular knowledge strategy, carefully selected and customised to the specific organizational context, has also been observed for firms in other industries (e.g. Prencipe and Tell 2001).

We thus adopt a dynamic and path-dependent view of dynamic capability development. A key assumption in this respect is that knowledge is a construct that accumulates and diminishes over time. Therefore, each type of knowledge defined by Zollo and Winter is susceptible to accumulation as well as obsolescence and depletion. Regarding the depletion effect: employee turnover and cognitive limitations related to the retention in short-term memory lead to a loss of experiential as well as articulated knowledge, since both kinds of knowledge reside in the heads of the employees. Codified knowledge will deplete through the redundancy of available information and the obsolescence of knowledge contained in the artifacts, due to environmental dynamism in its institutional, technological and competitive dimensions and the costs related to the maintenance and updating of the codes.

The rate with which these knowledge levels grow partly depends on dedicated investments in articulating knowledge (e.g. mentoring systems, de-briefing processes) and codified knowledge (e.g. in information systems, manuals). These articulation and codification efforts reflect the level of intentionality in the learning process (Zollo and Winter 2002). Investments in articulation and codification take the form of time and resources spent and, as such, may directly influence the time (and resources) available for direct exposure to events that trigger experiential knowledge[1].

Dynamic capability arises from the interplay between experience accumulation from the enactment of operating routines, and experiential, articulated and codified forms of knowledge. We conceptualize it as a set of related stock variables that accumulate or deplete over time as result of in- and outflows of knowledge related to processes of building, integrating or reconfiguring operating routines (cf. Dierickx and Cool 1989, Helfat and Peteraf 2003).

The key intuition relates to the problem that increasing levels of deliberate learning investments produce both positive and negative effects on the organization’s ability to adapt its operations. Positive effects have been connected to the development of higher levels of knowledge, particularly of causal nature, due to both the increasing levels of consciousness about the performance outcomes of prior experiences and the inquiries on its root causes and contingencies (Zollo and Winter 2002). In the case of codification, this 'consciousness effect' (see below) adds to the utility drawn by the use of the artifacts created to share agreed procedures and coordinate the execution of future routines. On the other hand, deliberate learning investments, especially if in the form of knowledge codification processes, comes with high costs related to the allocation of scarce resources (managerial time and attention, primarily), as well as with increasing levels of organizational inertia connected to the role of artifacts as institutionalized 'truth', which can become increasingly difficult to challenge.

As a possible explanation of this conundrum, we suggest the levels of articulation and codication of knowledge in the organization can be part of the distinction between the virtuous and the vicious effects of deliberate learning investments on the evolution of dynamic capabilities. At lower levels of articulated and codified knowledge, the virtuous effects will predominate, whereas the opposite effect may become increasingly dominant at higher levels of articulation and codification of knowledge. This argument is similar to Adler and Borys' (1996) distinction between the enabling and the coercive effect of bureaucracy. The key difference is that Adler and Borys identify specific organizational features potentially responsible for this crucial distinction, whereas we identify specific contingencies related to the stocks of different types of knowledge present at a given time in the organization.

We will therefore, first of all, distinguish between the effects of articulation and codification processes in the formal model that will be simulated. Second, we will consider explicitly, and jointly, the non-linear dynamics of both the positive and the negative effects of deliberate learning on the evolution of dynamic capabilities.

This will allow us to explore how variations in investments in deliberate learning strategies may lead to different and counterintuitive patterns of dynamic capability development and experience accumulation under different contingencies related to environmental dynamism.

Model Structure

The model was developed in Vensim and Ithink simulation software. A complete description of the model, including all equations, is given in the Appendix (attached as a separate document).

The model draws on a so-called capacitated delay structure (Sterman 2000). This structure arises when the impact of a set of related stocks (e.g. experience and knowledge) depends on the level of flows in and out of these stocks, but is also constrained by capacity boundaries (e.g. resource constraints).

Figure 1 provides an overview of the structure of the model. In this model, deliberate efforts to articulate and codify knowledge affect knowledge flows and levels, that in turn have an impact on dynamic capability, particularly in the form of the organization's ability to generate effective proposals for change in its operating routines. This ability to generate change proposals as well as environmental dynamism affect change in operating routines; moreover, environmental dynamism also influences knowledge development and attrition.

Figure 1 involves a stylized picture of the full model, that is described in more detail in the Appendix. In the diagramming notation in this figure, flow variables are depicted as pipes with valves. The rectangles represent the accumulated level of a particular variable (e.g. tacit experience). Clouds represent the sources for some flows and sinks for some outflows; these sources and sinks are outside the boundaries of the model. There are three inputs in the model that can be externally manipulated (e.g. in setting up experiments): articulation effort, codification effort, and environmental dynamism; the range of all three input variables is from 0 to 1.

The model draws on a number of key assumptions. First, we assume organizations are exposed to volatility in their product, market and technological environments. These types of dynamics are aggregated in the variable environmental dynamism (defined as a continuous variable from 0 to 1).

Second, we assume that dynamic capability arises from the interaction of tacit experience, articulated knowledge, codified knowledge and environmental dynamism, generating the ability to develop and implement change proposals. Dynamic capabilityis thus a second order quality of the entire system modeled here.

Operating routines (OR) are the key object of dynamic capability. Operating routines, embedded for instance in operating systems and procedures, change over time as a result of (a) inputs from either stable patterns of routine development or ad-hoc adjustments and (b) attrition. The development of operating routines occurs when a given group in the organization[2] produces change proposals in response to the environmental dynamism and opportunities the current routines are exposed to. Attrition of operating routines is a continuous process related to memory decay in the absence of frequent execution and possibly reinforced by increasing dynamism in the environment (e.g. e-banking that makes certain sales routines in the financial services industry less relevant).

As discussed in the previous section, tacit experience (TE) is an automated accumulation process in which people deal with their experiences in daily operations. The development of tacit experience therefore arises from engaging in operating routines, but may suffer from increasing environmental dynamism. This existing experience base is in fact exposed to a continuous process of attrition: some of the experience relevant today, will become obsolete tomorrow. Higher levels of environmental dynamism will reinforce this process.

The current base of tacit experience also feeds the process of knowledge articulation resulting in articulated knowledge (AK). The latter process, represented in the model as the Articulation rate, is a function of the effort made to articulate tacit experience (Articulation effort) and the available tacit experience (TE) divided by the current level of articulated knowledge (AK). Not all experience can be spoken. The TE/AK ratio therefore serves as an algorithm representing decreasing (articulation) returns to each additional effort to articulate tacit experience:

Articulation rate = ArticulationEffort * TE /AK

The development of codified knowledge (CK) is fundamentally different from knowledge articulation. Whereas each piece of tacit experience that is articulated leaves the stock of experience, codified knowledge can persist in its artifacts. In other words, knowledge that is effectively codified, can continue to be articulated, shared in face-to-face settings, and eventually adapted. In the model, codified knowledge is a function of the codification effort and the current pool of articulated knowledge (AK) divided by current stock of codified knowledge (CK). Not all articulated knowledge can be written and codified (Winter 1987, Kogut and Zander 1992, Cowan, David and Foray 2000); moreover, some articulated knowledge is deliberately not codified because of its sensitive and classified nature (Prencipe and Tell 2001). In this respect, the AK/CK ratio represents the decreasing nature of returns to codification efforts when CK is growing relative to AK (cf. Cowan and Foray 1997):

CK inflow = CodificationEffort * AK / CK

Figure 1: Overview of the Model

[pic]

In addition, codified knowledge may become obsolete, due to technological and market developments and changing product portfolios. Even if this kind of information is filed and retained, for example in a computer database or archive, we assume the stock of codified knowledge depletes when no one in the organization uses it or continues to contribute to its further development. Again, increasing environmental dynamism will reinforce the process of attrition of codified knowledge (cf. Figure 1).

Codification and articulation efforts use resources – for example, staff hours and attention – that cannot be allocated elsewhere in the organization (e.g. for client acquisition). In this respect, we assume the organization has a certain amount of staff resources available for three activities: knowledge articulation, knowledge codification, and operating routines. Resources spent in one of these three activities are not available to the other two.

The Development of Dynamic Capability

A key element of the model involves the role of knowledge articulation and knowledge codification in the evolution of dynamic capability. Based on the received literature, we identify four different effects of knowledge codification and articulation on the evolution of dynamic capability:

1. Tool Utility Effect: the first virtuous effect of knowledge codification follows from the usefulness of the tools that are produced in the process (Cowan and Foray 1997). These tools, often embedded in manuals, standard operating procedures, software applications and so on, help the organization to adapt its operating routines by enhancing the cognitive alignment among dispersed members of “how process X should be executed”, and facilitating the post-execution evaluation for eventual adaptation. For example, establishing a shared communication protocol facilitates the exchange of information and learning outcomes across individuals or groups (e.g. Dhanaraj, Lyles, Steensma and Tihanyi 2004) and allows the members of the groups to evaluate the effectiveness of their communication processes, eventually developing corrections and refinements. Holding environmental dynamism and other conditions constant, we will assume that increasing levels of knowledge codification will increase the organizational ability to generate change in operating routines.

2. Inertia Effect: when the level of codified knowledge becomes relatively high, compared to other forms of knowledge in the organization, this may stiffle the ability to produce effective proposals for changing operating routines. Codes represent the “way things should be done” and can discourage the challenge of the status quo, weakening the ability of the organization to adapt its operating routines. In the language of Leonard Barton (1992), increasing capability levels caused by investments in deliberate learning processes with highly inertial properties (i.e. knowledge codification, rather than articulation) can turn the same capabilities into rigities. We propose that the levels of knowledge codification present in the organization can be an important explanation of this, hitherto unresolved, puzzle. For example, Tripsas and Gavetti's (2000) describe Kodak’s failure to seize the opportunities from the digital revolution of photography. This failure might be explained, in addition to the role of cognitive inertia, by possibly high levels of accumulated codified knowledge, which in turn sustain strong but obsolete beliefs in the organization about the appropriate business model. Thus, we assume that increasing levels of codified knowledge produce a negative effect on the organizational ability to generate change in operating routines, and that this effect actually increases in strength with the rising stocks of codified knowledge.

3. Consciousness Effect: the process through which knowledge is articulated (e.g. team meetings, conversations between junior staff and their mentor) and codes are developed (e.g. in manuals, software and procedures) requires a significant amount of questioning about the causal linkages between actions and outcomes. Thus, the articulation and codification process therefore has a second, less obvious, virtue with regard to the organizational ability to adapt its operating routines. Members involved in articulation and particularly codification processes, in fact, are somewhat forced to ask questions – for example about the reasons for successes and failures in their prior experiences – thereby unveiling some of the causal ambiguity that covers most organizational activities (Zollo and Winter 2002). In a similar vein, knowledge articulation and codification raises the level of mindfulness about the effectiveness of its own processes (Weick, Sutcliffe and Obstfeld 1999) and draws attention to the needs to respond flexibly to contextual cues (Levinthal and Rerup 2006). We thus model the consciousness effect as an increasing level of dynamic capability at higher levels of articulated and codified knowledge. Note the effect of articulation will be of a lower magnitude, compared to codification, because the consciousness effect of codified knowledge is stronger in view of its focus on tangible artifacts (e.g. written protocols and manuals).

4. Experience Base Effect: this effect captures the conundrum that involves the implicit (negative) effect of codification process on the development of tacit experience, given the cognitive constrain introduced above. The more an organization invests in knowledge codification, the less time and resources it has to dedicate to the actual experiencing of interactions with the world. However, the development of tacit experience not only is a gradual cumulative process, but is also necessary to effectively codify, since it is the basis for the sense-making effort (Weick 1995, Cowan et al. 2000, Dhanaraj et al. 2004). For example, a firm's codified knowledge of a market is often facilitated through the kind of strong social ties that promote experiential learning between buyers and sales staff (Uzzi 1997). Thus, effective proposals for changes in operating routines need a substantial pool of tacit experiences to translate (e.g. written) proposals into actual development of the routines. Therefore, in line with the literature on implicit learning and consciousness (e.g. Cleeremans and French 2002), we assume a high level of tacit experience, relative to explicit knowledge, is required to maintain the consciousness effect referred to earlier. This assumption implies that a high ratio between explicit knowledge and tacit experience will decrease the ability to generate proposals for changes in operating routines.

One of the graphs in Figure 1 summarizes the consciousness and experience base effects, as a function of CK*w + AK*(1-w) divided by TE. The constant w represents the stronger impact CK has, compared to AK, on consciousness and the ability to generate change proposals – this constant is exogenous (set at 0.8) in the model. The consciousness effect involves a positive effect at lower levels of the ratio, with increasing marginal returns that subsequently turn into decreasing marginal returns. The idea that ad-hoc problem driven search provides a base kind of change capability is acknowledged here: if codified knowledge completely breaks down, the graph for the tool utility effect implies that the impact on the ability to change operating routines still is 0.2 (see the graph in Figure 1). The experience base effect involves a negative effect at high levels of the CK/AK ratio, with increasing marginal changes at first and decreasing marginal changes when the ratio moves toward 1. We assume an exogenously determined limit to the experience base effect, with a loss of 0.4 of the ability to generate change proposals.

The tool utility and inertia effects on the ability to generate effective change are assumed to be a function of the ratio between CK and AK. These effects are also summarized in Figure 1. The tool utility effect involves a positive effect at low levels of the CK/AK ratio, with increasing marginal returns that turn into decreasing marginal returns when this ratio moves to 1. We assume here that if codification is completely absent and the level of CK is thus zero, the net effect on the ability to generate change proposals is 0,2 – this represents the idea that a substantial pool of articulated knowledge creates an ability to change even without any knowledge codification (cf. ad-hoc kinds of search and problem solving). The inertia effect involves a negative effect of the same ratio at higher levels, with increasing marginal returns that turn into decreasing marginal returns when CK/AK is close to 1.

In sum, the ability to generate effective change proposals (GP) for transforming operating routines is defined as follows:

GP = ToolUtility&Inertia Effect * Consciousness&ExperienceBase Effect

= f [CK/AK ] * g [ {w*CK + (1-w)*AK} / TE ]

GP operates on the process of changing operating routines, reflecting the idea that a larger ability to generate high-quality change proposals will facilitate efforts to actually change routines to make them more effective in response to environmental cues. Moreover, resources are scarce, that is we assume there is a fixed amount of resources available in the organization to either execute operating routines or articulate and codify knowledge. (The amount of resources necessary to codify knowledge is considered significantly larger than in knowledge articulation processes.) As such, we can define change in operating routines as a function of the ability to generate effective change proposals (GP), environmental dynamism (ED), and resource constraints (RC):

OR development = (GP/ED) * RC

In sum, dynamic capability is thus conceived as a second order quality of the entire system. In addition to the effects of resource constraints and environmental dynamics, five feedback loops determine the evolution of organizational ability to change operating routines. The critical dimension of each of these five feedback loop arises from how tacit experience, articulated knowledge and codified knowledge affects either the Tool Utility-Inertia or the Consciousness-ExperienceBase effect on the ability to generate change in operating routines. Each feedback loop affects the ability to generate proposals and change operating routine development differently, in line with the assumptions discussed earlier.

Simulation Findings

The model outlined in the previous section can be used to simulate the evolution of deliberate learning, tacit experience accumulation and dynamic capability development. In particular, we explore how knowledge articulation and knowledge codificaton contribute to (or undermine) the development of dynamic capability, in response to increasing environmental dynamism. The model in the previous section serves to simulate a large number of different settings, characterized by (initial levels of) environmental dynamism, ability to generate change, and available resources. In this section we focus on knowledge/learning systems that have a low ability to generate change in operating routines at the outset, as a result of a relatively low consciousness level and a moderate tool utility effect (cf. Figure 1), in an environment that is relatively stable and under-resourced at the outset. This situation is often observed in small and medium sized (e.g. family-owned) firms that tend to react to rather than anticipate developments in their products, markets and technologies (e.g. Christensen 1997, Leonard-Barton 1992, Tripsas and Gavetti 2000).

To be able to capture short-term as well as long-term patterns, the model is simulated over a period of 200 quarters (or 50 years). In the simulation experiments that follow, the system begins in a steady state in which the inflow in each stock equals its outflow, environmental dynamism is 0.3 (on a scale from 0 to 1), and the articulation and codification effort is 0.05 respectively 0.05 – implying that the remaining 0.9 staff resources are allocated to operating processes.

To understand the behavior of the system in disequilibrium, the model was tested using a variety of changes in environmental dynamism, articulation effort and codification effort. This section describes a small but representative set of these simulation experiments.

The simulation run in Figure 2 is produced by exposing the system, in steady state conditions, to a 10% structural increase of environmental dynamism. In this simulation run, increased environmental dynamism directly affects the level of operating routine development, whereas the attrition in existing routines increases (see also Figure 1). Moreover, the (partial) breakdown in existing routines implies the development of tacit experiences is also affected; thus, the stock of tacit experiences diminishes. With the articulation or codification effort unchanged, the decreasing stock of experiences reduces the articulation rate, which in turn decreases the existing body of articulated knowledge (reinforced by an increasing attrition rate, that is higher anyway for articulated knowledge in the model). The net effect is that the organizational ability to change operating routines in response to new environmental imperatives is too low to cope with these imperatives. Therefore, the firm is not able to effectively develop new operating routines in response to environmental demands.

Figure 3 reports similar results of an experiment with a more substantial change in environmental conditions of 40 percent. This figure suggests that this more dramatic change in environmental conditions puts enormous strain on the system, undermining a major part of its operating routines.

Figure 4 shows the results of an experiment in which environmental dynamism increases by 50 percent. Under this additional strain, the system's operating routines and knowledge resources completely break down. Unlike the previous experiment the system cannot settle at a new steady state. These first three experiments suggest the model in this paper has a threshold, or tipping point, beyond which the behavior of the system in response to external impulses changes fundamentally (cf. Rudolph and Repenning 2002). Below the threshold the system is able to find a new steady state, but beyond the threshold vicious reinforcing feedback prevails and the system breaks down.

------------------------------------------

Insert Figure 2, 3 and 4 here

------------------------------------------

Figures 5 and 6 illustrates how a simple one-time additional effort to articulate or codify knowledge changes the behavior of the system exposed to the 10 percent increase in environmental dynamism (cf. Figure 2). Figure 5 shows the long-term effects of an additional effort to articulate knowledge during two subsequent quarters (starting half a year after the environment starts changing); this additional effort doubles the 0.05 structural effort made each quarter. Evidently, the additional effort invested in talking to each other (e.g. during strategy workshops, extra team meetings, etcetera) does not pay off, but rather reinforces the process pictured in Figure 2. This counter-intuitive result is mainly due to the loss in tool utility effect, that is not effectively compensated by the growing consciousness of the agents (see Figure 5).

By contrast, Figure 6 shows how effective an additional effort in codifying knowledge in this particular setting is. Doubling the codification effort in response to new environmental cues reinforces the tool utility effect, whereas the existing consciousness level is maintained by somewhat shifting the knowledge base. The difference between the patterns in Figure 5 and 6 is remarkable, and arises from the different impact of codified and articulated knowledge as well as the cumulative nature of the effects of articulation and codification interventions (cf. their path dependency).

-------------------------------------------

Insert Figure 5 and 6 here

-------------------------------------------

A similar set of experiments can be done in response to the more dramatic scenario of 40 percent change in environmental dynamism (cf. Figure 4). Figure 7 illustrates how the system behaves when the articulation effort is stepped up in two subsequent quarters. This gives similar patterns over time as for the response to the 10 percent change in environmental dynamism, described earlier.

In Figure 8 we tested the response to stepping up the codification effort. This result is strikingly different from the response to the same intervention in Figure 6. In this case, the system's change capacity diminishes, since the tool utility and consciousness effects created by knowledge codification fail to overcome the stronger inertial effects due to stronger dynamics in the environment. The patterns observed in Figure 8 hold for different sizes of the additional codification effort. This suggests that under more dramatic environmental change conditions, knowledge codification alone fails to support the firm’s ability to adapt.

The simulation run in Figure 9 shows how the system can deal quite effectively with the 40% increase in dynamism by stepping up both the codification and the articulation effort during two quarters. These findings suggest that high levels of environmental dynamism need to be dealt with by investing in both knowledge articulation and codification, rather than relying only on one of the two deliberate learning tools.

-------------------------------------------

Insert Figure 7, 8 and 9 here

-------------------------------------------

Discussion and Conclusions

This paper set out to study the mechanisms underlying the development of dynamic capability in firms and the impact of environmental dynamism on their relative effectiveness. The model proposed tries to be sensitive not only to the contrasting (positive and negative) effects related to deliberate learning investments, but also to the nuances between the role of different types of deliberate learning processes (knowledge articulation vs. knowledge codification) and to the trade-offs inherent in the choice of one or the other, or of none at all, that is relying on tacit knowledge accumulation processes.

The system simulated by the model shows, first of all, clear boundaries for the reliance on tacit knowledge accumulation processes, as environmental dynamism increases. This is an important result, given the strong emphasis in the literature on the virtues of tacit knowledge for organizational learning and adaptation processes (e.g. Nonaka and Takeuchi 1995, Dhanaraj et al. 2004). Whereas tacit experience does have a role to play in determining the firm’s ability to adapt, prior work may not have given sufficient consideration to the fact that the usefulness of tacit kinds of knowledge is particularly sensitive to the stability of the environmental conditions.

A second implication of the results of simulation runs is that investments in deliberate learning effects are not a homogeneous category, that is their effects on the development of dynamic capability differ in kind, not only in magnitude. More precisely, codification seems to be a much more effective learning and adaptation strategy at intermediate levels of environmental dynamism.

At high levels of dynamism, though, even a knowledge codification strategy (by itself) shows clear limitations. The inertial effects produced by the artifacts developed in the codification process become powerful enough to counteract the positive effects on the ability to develop higher causal knowledge (consciousness effect, in particular). In these contexts, similar to those described by Eisenhardt (1989) in her work on high velocity environments, knowledge articulation regains the upper hand when combined with codification efforts. Intuitively, articulation can strike the appropriate balance between the need to penetrate causal ambiguity and the pressure to reduce the inertial effects of codification processes. This result can also be understood in terms of the notion of 'simple rules' that Eisenhard and Sull (2001) describe. Simple rules do require a (low) dose of knowledge codification processes, and a (more abundant) dose of investment in knowledge articulation to figure out how to cope with complexity, eventually adapting the few rules the firms evolves on.

Putting it all together, the managerial insight that emerges from these results is that it is crucial for any firm to understand how to adapt its deliberate learning approach to the environmental conditions it is facing. The relative effectiveness of the two deliberate learning processes changes, in fact, quite radically depending on whether the firm faces an intermediate or a high level of dynamism.

Despite this set of contributions to the received literature, there are several simplifying assumptions we had to make in this initial simulation effort, which correspond to future research work to be done. First, we have not modeled the characteristics of the tasks that identify the operating routines in eventual need of adaptation. Zollo and Winter (2002) develop theory on some of these characteristics and their impact on the relative effectiveness of deliberate learning mechanisms (homogeneously considered). One possible way to expand on the results of our model is to consider explicitly the influence of task frequency, heterogeneity and causal ambiguity on the way the environment affects the optimal learning and adaptation strategy.

A second set of contingencies that has been left out of the analysis has to do with firm-level effects, such as its size, structural form, strategic posture and overall performance, just to name a few of the most important ones. It would be very important, to further develop our current understanding of how organizations learn to adapt, if we could integrate their own structural features in the model.

Of course, putting task effects, firm effects and environmental effects in the same simulation model would produce results that could be difficult to analyze, let alone understand. That is where simulation approaches might be leveraged to guide empirical inquiry, the ultimate test of validity and generalizability of the theoretical effort. We certainly look forward to the future development in at least some of these directions and hope that this particular paper serves to stimulate future scholars’ curiosity toward the next steps in our quest to understand how organizations learn to change and adapt.

References

Adler, P. S. & B. Borys. 1996. Two types of bureaucracy: Enabling and coercive. Administrative Science Quarterly, 41, 61-89.

Christensen, C. 1997. The Innovator's Dilemma. Harvard Business School Press, Boston, MA.

Cleeremans, A. & R. M. French (eds.) 2002. Implicit Learning and Consciousness: An Empirical, Philosophical and Computational Consensus in the Making. Psychology Press, Brighton, NY.

Cowan, R. & D. Foray. 1997. The economics of codification and the diffusion of knowledge. Industrial and Corporate Change, 6, 595-622.

Cowan, R., P. David & D. Foray. 2000. The explicit economics of knowledge codification and tacitness. Industrial and Corporate Change, 9, 211-253.

Dhanaraj, C., M. A. Lyles, H. K. Steensma & L. Tihanyi. 2004. Managing tacit and explicit knowledge transfer in IJVs: The role of relational embeddedness and the impact on performance. Journal of International Business Studies, 35, 428-442.

Dierickx, I. & K. Cool. 1989. Asset stock accumulation and sustainability of competitive advantage. Management Science, 35, 1504-1511.

Eisenhardt, K. M. 1989. Making fast strategic decisions in high-velocity environments. Academy of Management Journal, 32, 543-576.

Eisenhardt, K. M. & J. A. Martin. 2000. Dynamic capabilities: What are they? Strategic Management Journal, 21, 1105- 1121.

Eisenhardt, K. M. & D. N. Sull. 2001. Strategy as simple rules. Harvard Business Review, 79 (1), 107- 116.

Gavetti, G. 2005. Cognition and hierarchy: Rethinking the microfoundations of capabilities' development. Organization Science, 16, 599-617.

Gavetti, G. & D. Levinthal. 2000. Looking forward and looking backward: Cognitive and experiential search. Administrative Science Quarterly, 45, 113-137.

Hansen, M.T., N. Nohria & T. Tierney, T. 1999. What's your strategy for managing knowledge? Harvard Business Review, 77 (2), 106-116.

Helfat, C. E. 1997. Know-how and asset complementarity and dynamic capability accumulation: The case of R&D. Strategic Management Journal, 18, 339-360.

Helfat, C. E., S. Finkelstein, W. Mitchell, M. Peteraf, H. Singh, D. Teece & S. Winter. 2006. Dynamic Capabilities: Understanding Strategic Change in Organizations. Blackwell Publishing, Oxford (UK), in press.

Helfat, C. E. & M. A. Peteraf. 2003. The dynamic resource-based view: Capability lifecycles. Strategic Management Journal, 24, 997-1010.

Helfat, C. E. & R. S. Raubitschek. 2000. Product sequencing: Co-evolution of knowledge, capabilities and products. Strategic Management Journal, special issue, 21 (10-11), 961-979.

Kogut, B. & U. Zander. 1992. Knowledge of the firm, combinative capabilities, and the replication of technology. Organization Science, 3, 383-397.

Leonard-Barton, D. 1992. Core capabilities and core rigidities: A paradox in managing new product development. Strategic Management Journal, Summer special issue, 13, 111-126.

Levinthal, D. & C. Rerup. 2006. Crossing an apparent chasm: Bridging mindful and less-mindful perspectives on organizational learning. Organization Science, 17, 502-513.

Nonaka, I. & H. Takeuchi. 1995. The Knowledge-Creating Company. Oxford University Press, Oxford.

Prencipe, A. & F. Tell. 2001. Inter-project learning: Processes and outcomes of knowledge codification in project-based firms. Research Policy, 30, 1373-1394.

Rudolph, J. W. & N. P. Repenning. 2002. Disaster dynamics: Understanding the role of quantity in organizational collapse. Administrative Science Quarterly, 47, 1-30.

Sterman, J. D. 2000. Business Dynamics: Systems Thinking and Modeling for a Complex World. McGraw-Hill, Boston.

Teece, D. J., G. P. Pisano & A. Shuen. 1997. Dynamic capabilities and strategic management. Strategic Management Journal, 18, 509-533.

Tripsas, M. & G. Gavetti. 2000. Capabilities, cognition, and inertia: Evidence from digital imaging. Strategic Management Journal, 21, 1147-1161.

Uzzi, B. 1997. Social structure and competition in interfirm networks: The paradox of embeddedness. Administrative Science Quarterly, 42, 35-67.

Weick, K. E., K. M. Sutcliffe & D. Obstfeld. 1999. Organizing for high reliability: Processes of collective mindfulness. Research in Organizational Behavior, 21, 81-123.

Weick, K. E. 1995. Sensemaking in Organizations. Thousand Oaks, Sage.

Winter, S. G. 1987. Knowledge and competence as strategic assets. In: D. Teece (ed.), The Competitive Challenge: Strategies for Industrial Innovation and Renewal. Ballinger, Cambridge.

Winter, S. G. 2003. Understanding dynamic capabilities. Strategic Management Journal, 24, 991-995.

Zollo, M. & S. G. Winter. 2002. Deliberate learning and the evolution of dynamic capabilities. Organization Science, 13, 339-351.

Zott, C. 2003. Dynamic capabilities and the emergence of intraindustry differential firm performance: Insights from a simulation study. Strategic Management Journal, 24, 97-125.

Figure 2: The development of AK, CK, TE, OR and the ability to change OR: simulation results of a 10% structural increase in Environmental Dynamism (system starts in steady state).

[pic]

Figure 3: The development of AK, CK, TE, OR and the ability to change OR: Simulation results of a 40% structural increase in Environmental Dynamism (system starts in steady state).

[pic]

Figure 4: The development of AK, CK, TE, OR, and the ability to generate change in OR: Simulation results of a 50% structural increase in Environmental Dynamism (system starts in steady state).

[pic]

Figure 5: Stepping up the Articulation Effort in quarter 12 and 13, in response to Environmental Dynamism increasing 10% (compare OR with Figure 2).

[pic]

Figure 6: Stepping up the Codification Effort in quarter 12 and 13, in response to the 10% Environmental Dynamism increase (compare OR with Figure 2).

[pic]

Figure 7: Stepping up the Articulation Effort in quarter 12 and 13, in response to 40% increase in Environmental Dynamism (compare OR with Figure 3).

[pic]

Figure 8: Stepping up the Codification Effort in quarter 12 and 13, in response to 40% increase in Environmental Dynamism (compare OR with Figure 3).

[pic]

Figure 9: Stepping up both the Articulation and Codification Effort in quarter 12 and 13, in response to 40% increase in Environmental Dynamism.

[pic]

-----------------------

[1] This reflects the off-line versus on-line distinction made by Gavetti and Levinthal (2000) in their study of search in strategy-making processes.

[2] Note that the group proposing changes to a given routine does not need to be one that executes it. For example, manufacturing routines are usually adapted by a specialized group responsible for technical improvements (process innovation unit), which is the holder of the dynamic capability.

-----------------------

[pic]

Tool Utility effect

Inertia effect

[pic]

Consciousness effect

Experience Base effect

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download