The Strength of Corporate Culture and the Reliability of Firm Performance

The Strength of Corporate Culture and the Reliability of Firm Performance

Transcript of The Strength of Corporate Culture and the Reliability of Firm Performance

Abstract

Paper is consider the relation between the exploration of new possibilities and the exploitation of old certainties in organizational learning. It examines the complication in allocating resources between two, particulary those introduced by the distribution of cost and benefits across time and space, and the effects of ecological interaction. Two general situation involving the development and use of knowledge in organization are modeled. 1st, case of mutual learning between members of an organization and an organizational code.  2nd, case of learning and competitive advantage in competition for primacy. The paper develops an argument that adaptive process, by refining exploitation more rapidly than exploration are likely to become effective in the short run but self-destructive in the long run.

The possibility that certain common organizational practices ameliorate that tendency is assessed.

Exploration: purpose of discovery the resouces or information

--> Search, Variation, Risk Taking, Experimentation, Play, Flexibility, Discovery, Innovation.

Exploitation: Using something in an unjust

-->Refinement, Choice, Production, Efficiency, Selection, Implementation, Execution.

Adaptive system: Engage in exploration to the exclusion of exploitation are likely to find that they suffer the costs of experimentation without gaining many of its benefits. Therefore, as a result maintaining an appropriate balance between exploration and exploitation is a primary factor in system survival and prosperity. Organization make explicit and implicit choice between two: Explicit choices in calculated decision about alternative investments and competitive strategies. Implicit choices are buried in many features of organizational forms and customs ( organizational procedures--accumulating and reducing slack, in search rule and practive in the way in which targets are set and changed and in incentive system.

Theories of Organizational Action:

In rational model of choice-->balance btw E&E is discussed classically in terms of a theory of rational serach (alternatively, probability).

In theories of limited rationality--> Discussion of the choice btw E&E emphasize the role of targets or aspiration levels in regulating allocations to search.

In studies of Organizational Learning-->Problems of balancing E&E is exhibited in distinctions made between refinement of an existing technology and invention of a new one.

In evolutionary model of organizational forms and technologies-->Discussion of the choice btw E&E are framed in terms of balancing the twin process of variation and selection.

The Vulnerability of Exploration

Compared to returns from exploitation, and exploration are systematically less certain, more remote in time and organizationally more distant from the locus of action and adaption. 

The Social Context of Organizational Learning

Contain 2 distinctive features of the social context:

1. Mutual learning of an organization and individual. Organizationas store knowledge in their procedures, norm, rules and forms. They accumulate such knowledge over time, learning from their members. At the sametime, individual in an organization are socialized to organizational beliefs
2. The context of competition for primacy.Organizational often compete with each other under conditions in which relative position matters.

Mutual Learning in the Development of Knowledge

A model of Mutual Learning - process begins with an organizational code characterized by neutral beliefs on all dimensions and set of individuals with varying beliefs that exhibit on average, no knowledge. 

Basic Properties of the Model in a Closed System

Effect of learning rate: higher rates of learning lead, on average, to acheiving equilibrium (As individual in the organization become more knowledgeable, they also become more homogeneous with respect to knowledge). Effect of learning rate heterogeneity: effect on improving in organizational knowledge and on long-term individual improvement suggest that the advantage to having mix of fast and slow learners in an organization. 

Basic Properties of the Model in a More Open System

Effect of personel turnover: Alternative way to producing variability in an organization. The greater the turnover, the shorter the average length of service and the lower the average individual knowledge at any point. Effects of environmental turbulence: useful on turnover, but it produces a disparity btw code knowledge and average knowledge of individuals in the organization. 

Knowledge and Ecologies of Competition

• Competition and the Importance of Relative Performance
• Learning, Knowledge and Competitive Advantage
• Competition for Relative Position and Strategic Action

Litter Model and Old Wisdom

Learning, analysis, imitation, regeneration and technological change are major components of any effort to improve organizational performance and strengthen competitive advantage. Each involves adaptation and a delicate trade-off btw E&E. The present argument has been that these trade-offs are affected by their context of distributed costs and benefits and ecological interaction. The essence of exploitation is the refinement and extension of existing competences, technologies and paradigms. Its returns are positive, proximate and predictable. The uncertain, distant and often negative. Thus, the distance in time and space btw the locus of learning and the locus for the realization of returns is generally greater in the case of exploration than in the case of exploitation, as is the uncertainty. Model suggest on consideration involved in thinking about choice between E&E and in sustaining exploration in the face of adaptive processes that tend to inhibit it.

The Strength of Corporate Culture and the Reliability of Firm Performance

Abstract

Prevailing research claims that strong corporate cultures improve firm performance by facilitating internal behavioral consistency. Addresses an unexamined implication of this argument by analyzing the effect of strong corporate cultures on the variability of firm performance. The relationship depends on how strong cultures affect organizational learning in response to internal and external change.

Paper hypothesize that strong culture firms excel at incremental change but encounter difficulties in more volatile environment. Results of analyses of a sample of firms from a broad variety of industries show that in relatively stable environments, strong culture firms have more reliable performance. In volatile environments, however, the reliability benefits of strong cultures disappear.

Corporate Culture

Strong-culture firms have encountered great difficulties in responding to changes in their environment. Performance variability, an important outcome in its own right because it plays a central role in a variety of theoretical approach to organizations. (high viarable performance may increase the frequency of risk taking behavior. Organizational ecologists have attributed causal importance to performance variability by arguing that external stakeholders typically attach value to predictable performance, giving reliable firms a survival advantage. Behavioral theories, risk taking by managers depend on firm performance relative to aspiration level.

The strength of corporate cultures and firm performance

Culture is a measurable characteristic of organizations. A system of shared values and norms that define appropriate attitudes and behavior for organizational members. A culture can be considered strong if those norms and values are widely shared and intensely held throughout the organization.

Organization culture define as: 

• A social control mechanism
• It framed people interpretations of organizational events and basic assumptions about organizational processes
• It provide group members with a way of giving meaning to their daily lives, setting guidelines and rules for how to behave, and most important, reducing and containing the anxiety of dealing with an unpredictable and uncertain environment.

Performance benefits of strong cultures

Widespread consensus and endorsement of organizational values and norms facilitates social control within the firm. when there is broad agreement that certain behavior are more appropriate than others, violations of behavior norms may be detected and corrected faster. 

Enhance goal alignment. With clarity about corporate goals and practices, employees face less uncertainty about the proper course of action when faced with unexpected situations and can react appropriately. Enhance employees motivation and performance because they preceive that their action are freely chosen. Performance reliability depends on 2 factors: the consistency with which a firm perform its organizational routines and the degree to which those routines are well adapted to changing environmental conditions.

Culture, Learning, and Performance Variability

--culture ultimately reflects the groups effort to cope and lean and is the residue of learning process.

Hypothesis 1: Firms with strong corporate cultures will exhibit more reliable (less reliable) performance. Knowledge makes performance more reliable.As work is standardized as techniques are learned, variability,both in the time required to accomplish tasks and in the quality of task performance is reduced.

Culture, Environmental Volatility, and Performance

--Organizational performance in changing environments depends on the ability of the firm to modify its routines in response to changes in conditions.

Hypothesis 2: As industry volatility increases, the positive effect of culture strength on mean performance declines.

Hypothesis 3: As industry volatility increase, the positive effect of culture strength on performance reliability declines.

Strong culture organizational should have greater difficulty responding to environmental volatility than weak-culture organizational. If environmental change sharply reduces the value of the organization existing routines, strong culture firms should have greater difficulty regaining their footing

Hypothesis 2: that the positive effect of culture strength on mean performance should decrease in volatile environments was not supported. Because strong culture firms are in a relatively good position to weather short-term volatility, it may only be when there are truly radical shifts in the underlying technologies and competitive conditions in an industry that the mean performance benefits of a strong culture disappear.

Conclusion

Firms with strong cultures incur a tradeoff with respect to their adaptive ability in the face of environmental change.

Strong corporate cultures facilitate reliable performance in relatively stable environmentsm but as volability increases, these benefits are dramatically attenuated. This pattern is consistent with the fundamental tradeoff between exploration and exploitation noted by March (1991) and suggests that strong culture firm excel at exploring and discovering new competencies that better suit changing environmental conditions.

It still depends on the existence of environmental conditions that reward exploration. If the environment changes to reward efficiency exploration and exploitation of organizational routines, firms that are strongly committed to exploration should have greater difficulty adapting than firms with weak cultures of exploration, because such an environment demands relative stability in organizational routines and the ability to make incremental improvement in efficiency. 

It create competitive advantage not only by increasing motivation and facilitating coordination and control, but also by leaving the strong culture firm in a stronger position to respond to invesment opportunities that might solidify its competitive advantage.

Exploration and exploitation in organizational learning

Exploration and exploitation in organizational learning

Citation: James G. March (1991) Exploration and exploitation in organizational learning.

Summary:

Exploration and exploitation in organizational learning is simulation based paper on organizational learning. It framed around the idea of exploring new possibilities versus exploiting existing ones but focuses primarily on two sets of models of organizational learning.

The paper first introduces what it calls the "exploration / exploitation trade-off" which is connected to a wide variety of existing innovation literatures on product versus product innovation, on dominant designs, and even to the role of targets and aspirations from Cyert and March. There are limited resources which will be devoted to either exploring new possibilities or to exploiting existing ones. March uses this as the framing for his discussion of learning processes.

Marches first set of models is a model of mutual learning with an organization. Essentially, he assumes that organizations learn routines based on the knowledge of its members and that members also learn from the organization through a process of socialization. He conclusion is that, "the gains to individuals from adapting rapidly to the code (which is consistently closer to reality than the average individual) are offset by second-order losses stemming from the fact that the code can learn only from individuals who deviate from it."

March models both a closed system and a more open system. In a turbulent environment (i.e., one in which the ground truth is shifting) organizations with any socialization are doomed unless turnover is introduced where new ideas can come in and keep the organization on at least part of a search path.

March's second model attempts to capture the effect of competition between organizations on learning and shows an important effect of competition. He explains that, "as the number of competitors increases, the contribution of the variance to competitive advantage increases until at the limit, as N goes to infinity, the mean becomes irrelevant."

Marches takeaways are connected to "old wisdom." Adjustment to codes within organization is a threat to learning that can have bad effects. In general, "the returns to fast learning are not all positive, that rapid socialization may hurt the socializers even as it helps the socialized, that the development of knowledge may depend on maintaining an influx of the naive, and ignorant, and that the competitive victory does not reliably got to the properly educated."

Theoretical and practical relevance:

Marches article has been cited more than 6,000 times and is one of the most highly cited articles in the organizations literature. It is a critical piece in the literature on organization learning. Moreover, it is a go-to citation for anybody talking about exploration and exploitation -- a framing which has become key and frequently present and has even found its way into the titles of hundreds of papers.

Measuring Knowledge Exploitation and Exploration

Measuring Knowledge Exploitation and Exploration: An Empirical Application in a Technological Development Center in Brazil

Silvio Popadiuk 1, Marcos Antonio Franklin 2 Patricia Gonçalvez Vidal 3, Lilian Aparecida Pasquini Miguel 4 y Vanderli Correia Prieto 5

Recibido: 28-01-2010 - Aprobado: 01-05-2010

---

ABSTRACT:

This article presents a literature review on the concepts of exploration and exploitation of knowledge. From this review, it is showed that the concepts of exploration and exploitation can be analyzed through internal or external environments perspective, related to seven dimensions, namely: knowledge, innovation, strategy, costs, efficiency, competition, and partnership. From this theoretical conception, a closed and structured questionnaire was developed, that contained attributes related to the dimensions above. These attributes were assessed through a six-points Likert scale. This questionnaire was applied in a research and development institute of a major multinational firm in the telecommunication sector, in Brazil. The main purpose of this article is to assess the fit between what have been proposed in the literature on exploitation and exploration and what have been practiced in the corporate world. The results revealed two distinct groups of employees in this institute, regarding how these employees perceived their activities to be innovative. Regarding these two groups, there are significant differences regarding how these two groups evaluated the attributes related to the exploration and exploitation of knowledge.

Keyword: Exploration, Exploitation, Fit, Innovation, Brazil

Introduction

The process of decision making to invest resources in an organization has been the key issue for over decades of organizational studies (March and Simon, 1958; Cyert and March, 1963). Lately, one of the focuses of organizational theorists is to understand how to invest the organizational resources in a way to balance what March (1991) has called exploration of new knowledge and exploitation of available knowledge (Dosi and Marengo, 1997; Rothaermel, 2001). The organization may choose to invest in the refinement of an existing technology or on any other activity that is oriented toward increasing efficiency (Dosi and Marengo, 1997) or it may choose to invest in activities that search for novel processes or the discovery of new opportunities.

March (1991) argues that the organizations that engage in exploration to the exclusion of exploitation are likely to be penalized by the experimentation without gaining many of its benefits, whereas the organizations that engage in exploitation to the exclusion of exploration are likely to find themselves trapped in a suboptimal equilibrium. Companies thus face a trade-off between exploration and exploitation: If the company does not reach an appropriate equilibrium between the two ends of the continuum, the company may end up suffering from sub-optimization. In sum, firms face a trade-off between the creation of new capabilities (exploration) and the development of current capabilities (exploitation).

This article has the following research question: what is the fit between the theoretical considerations regarding exploration and exploitation and the activities being developed and used in a managerial setting? From this research question, this article has five purposes: to present a literature review of exploration and exploitation of knowledge; to characterize seven implicit dimensions regarding the theoretical concepts of exploration and exploitation of knowledge; to present an exploratory questionnaire using attributes of exploration and exploitation of knowledge; to apply this questionnaire in a pilot project inside a research and development institute of a major multinational firm in the telecommunication sector working in Brazil; and to elaborate analytical considerations regarding the fit or lack of fit regarding the theoretical context and the organizational context.

This article is structure as follows: next topic presents the strategies of exploitation and exploration of knowledge (March, 1991), summing up to the seven conceptual dimensions identified from this literature review. The second part introduces the methodology used to develop the scale as well as its tests; the third part describes the results. The final part, presents the discussion of the findings as well as the limitations of the study and the possible contributions both to practice and to the academia.

Literature review

Exploration and Exploitation

March (1991), in his seminal article, when looking from an organizational learning point of view, argues that there is a relationship between exploration and exploitation in the adaptive process in the organizations. Exploration refers to the actions and activities related to research, search, risk, experimentation, playing, flexibility, discovery, and innovation. Exploitation refers to refinement, choice, production, efficiency, selection, implementation and execution.

The essence of exploitation is the refinement and extension of the existing competencies, technologies, paradigms; a refinement that will lead to returns that are positive, near and predictable. While exploration is the experimentation of new alternatives, with returns that are far away on time, uncertain, and usually negative (March, 1991). March (1991) argues that the organizations make explicit or implicit choices for one or other, since the resources are scarce and the difficulties associated with the choices lead to complications in specifying the appropriated trade-offs.

The adaptable systems related to exploration which are excluded from the exploitation allegedly suffer from the costs related to the experimentation, without obtaining most of its benefits. Inversely, the systems that are related to the exploitation which are excluded from exploration are in a stable sub-optimal equilibrium (March, 1991). The organization choice for the exploration or exploitation resides, according to March (1991), on the rational search theory.

March (1991), cited Radner and Rothschild (1975) and Hey (1982), to explain that there are several possibilities to invest, which one characterized by the likelihood of distributing unknown. As time passes, new information about these distributions is accumulated, but the choices should be done among investing more in new alternatives and improve the future returns, or to improve the present returns. March (1991) points out that the studies on organizational learning direct the choice between exploration and exploitation regarding the search for new technologies or the refinement of existing technologies. The alterations in the environment are also pointed out since these changes need the generation of new alternatives to the survival of the firm. Due to the environmental turbulence, organizational diversity and competitive advantage, the evolutionary dominance of one organizational practice is sensitive to the relation between the variation related to the exploration reflected by the practice and by the change rate in the environment.

The learning process is reinforced, either positively or negatively, as a result of the path chosen by the organizations, consequently the choice between two alternatives is path dependent (March, 1991). To March (1991), the organizations accumulate experience in its norms, routines and forms at the individual and organizational level.

Teece, Pisano and Shuen (1997) also argue that the future position the organization may reach is a function of its current position, a position that suffers influence of the path chosen before, and the path ahead of it. This is due to the investment made a priori as well as the routine repertory. Teece et al. (1997) also posit that learning is a process of attempts, returns and evaluation. This way, the path dependency is amplified when there are conditions to the increase of returns related to the adoption of a technology. The products and the technologies are more attractive as they are adopted, characterized as a demand phenomenon.

The increase in returns can have different sources, as network externalities, the presence of complementary assets, and the infra-structure support, learning by doing and scale economies of production and distribution. Since there are several sources to increase returns, the anterior position of the firm may affect the capacity to exploit increasing returns.

The trade-off between exploration and exploitation occurs simultaneously and involve conflicts at short and long run, and it also has gains to individual and collective learning. To Gilsing (2002), there is a social construction of knowledge related to the choice between exploration and exploitation: the resulting learning is formed by new ideas, changes in beliefs, new and adapted institutions, change in the design of the structure of the networks and the adaptation of coordinating mechanisms.

Gilsing (2002) posit that when there is innovation, there is a rupture of the routine, there is a discontinuity that shows the beginning of a new base of knowledge that is totally tacit. In a sectorial system, the tacit nature of knowledge pushes the participants to approximate in a way to obtain a reduction in the cognitive distance, through the near interaction. There is a surge of varieties of ideas, prototypes, and demos. This variety is specially formed from the results in the learning processes and it consists of new knowledge highly tacit, and very specific to a limited number of firms. The environment, in phase of exploration, is characterized by low selection through the coordinating mechanisms, like the social norms and reputation; also there is a large variety and learning results that affect mainly the new base of knowledge (Gilsing, 2002). As a consequence, firms start to focus on the fit between demand needs and new markets for the new knowledge base. Also, the competition increases once the new knowledge base is already extensive to allow the development of several types of designs through several networks of exploration, creating competition among them. The demand and the combination of social and technical norms reinforce directly the activities of search and, by doing it, create limits to the new knowledge base. The results from the learning, such as new ideas and knowledge, should be adjusted to the emerging selective environment. A process of stability begins when a dominant design is chosen and there is the elimination of competing designs (Utterback, 1994). As a result, the learning focuses the change from exploration to exploitation. The competition intensifies and, usually, creates an incentive to focus on cost reduction. From this moment, the technical norms and formal procedures lead the firms to the search for the “scale effects” (Gilsing, 2002), i.e., the economic search for exploiting economy of scale gains.

Li, Schoenmakers and Vanhaverbeke (2006) evaluated the trade-off between exploration and exploitation from the learning experiences, and observed that, in a context of Schumpeterian competition, the returns would be diluted through imitation and destructive innovation. Li et al. (2006) argue that exploration and exploitation are activities related to the search of knowledge. In a context where a given performance goal must be reached, the firms need to find solutions to reach the goal. On the other hand, if the current performance is better than the expected one, so the organization enter in a calm path, and it encourage the search for activities that are not justifiable in terms of expected returns (Li et al., 2006). Lundvall (1999) points as a good plan to firms that want to increase learning to combine indoor activities that are dissimilar and complementary.

Lundvall (1999) argues that if a firm tries to create a disequilibrium between what it can do and what it is required from it to accomplish its task are good way to stimulate the learning as well as to increase the set of competencies of the firm. There are three categories related to the basic functions of the firm: (1) allocate scarce resource, (2) exploit sub-utilized resources when entering new activities and (3) speed up the learning and the creation of new competencies (Lundval, 1999, p. 12). The firm will reallocate its resources if there is an alteration of the price of factors. It will exploit the un-utilized resources and will use its existing knowledge base in connection to the introduction of new products. Its success will depend upon the construction of new competencies.

The learning reasons are divided between exploration and exploitation. Exploration is directly connected to the tacit knowledge that is assumed to be a pre-condition to the development and acquisition of new knowledge. Learning in exploitation is related to the explicit knowledge that can be applied to refinement or improvement of current knowledge (Karlson, 2005).

Li et al. (2006) posit that the activities related to the knowledge search have several dimensions one of them is the timing. Timing Exploitation is the creation of new knowledge through the search of recent knowledge and timing exploration is the creation of new knowledge through the search of knowledge far away in time. Considering the limited rationality and the path dependency, individuals tend to search for alternatives that are related to their current knowledge, and valuable choices are lost in time. Also, the choice that has been made was done in lie of complementary assets that were not available at that moment.

Another dimension is the geographic or institutional (Li et al., 2006) which is related to having the resources in common available at the same geographic area. This proximity facilitates the transfer of knowledge which in an exploration phase is highly tacit and can be transfer more easily through interactions and practices.

Li et al. (2006) also consider that the activities involved in exploration and exploitation are important to the process of innovation. They point out that the process of innovation involves stages of search of knowledge, knowledge recombination, finding an innovative solution and its commercialization (Li et al., 2006, p. 11). The authors argue that starting from an application of one technology until its expansion, problems may accumulate in the process of differentiation of this technology and the exchange of relevant practices up to the point where efficiency and returns diminish considerably, At this point, there is no possibility to generate additions or modifications. This is the necessary incentive to search for new structures or to generate the Schumpterian creative destruction.

Teece et al. (1997) advances the notion of competitive advantage through the exploration of possibilities that are external and specific to the firm, as well as the development of new opportunities (Penrose, 1959; Teece, 1986; Wernerfelt,1984), by developing and explaining the term dynamic capabilities, to strategic manage the capability of adapting, integrating, reconfiguring abilities, resources and functional competencies to the environment, specially since the business environment is always changing. Also, the dynamic capabilities are complemented by the path dependency, i.e., the firms in several points of their trajectories make choices that are irreversible, investing in one competency in lieu of another competency. These choices will determine not only the choices presented today but also in the future. The capability that the firm has should be compatible with the user needs, unique and difficult to be replicated. Teece et al. (1997) complements stating that the firm´s current capability could be the base of the diversification to new markets as well as new products. However, these dynamic capabilities can be obtained only through the internal development of such a capability.

A valuable consideration to the exploration and exploitation discussion is that the core competency and capacity of the firm are involved in the organizational processes. These processes are molded by the firm´s assets (internally or in the market), and through the evolutionary path that the firm has adopted (Teece et al., 1997). Gilsing (2002) has also considered the characteristics of the co-evolutionary path, describing it by the search to predict which variables in the system answer to changes in other variables in the system or changes in the system itself.

Masini, Zollo and Wassenhove (2004) argues that the strategic decision between product differentiation and cost leadership lead to a trade-off between exploration and exploitation, since if the organization decides to compete on the differentiation, the organization will worry about its products being copied or imitated. Therefore, the costs associated with this strategy will be higher. Following the logic, the organization that competes on cost leadership is not worried about the risk of having its product being copied or imitated.

Adner (1999) argues that the tension between exploration and exploitation is not axiomatic, i.e., the exploration of new technologies does not prevent their exploitation in the market. The firms could influence the investment in activities in the market to speed up the process of exploration through the process of exploitation by focusing on the needs heterogeneity as well as the sources of retro-feeding in the market. Adner (1999) posits that the technological development can happen inside the boundaries of the firm and can be structured to influence the market opportunities. The idea here is to demonstrate the possibility that technology development and its commercialization can guide themselves mutually. When dealing with emerging technologies, the firms can exploit the market diversity by allowing the target market to co-evolve with the state of its technology.

Prietula and Weingart (2005) used the terms exploration and exploitation in describing models of negotiation. Exploration reflects a more risky and uncertainty component of the joint search of two negotiators, where they try to define, refine, and align their knowledge to establish a common base between them. This way, the authors define exploration as a process with wide supply of potential samples. In one hand, they suggest that exploitation is an activity much more coordinated, that emerge from the common base of knowledge which is created when the two parts converge to an understanding of what is the possible solution for the negotiation.

Ahuja e Katila (2002) used the concepts of exploration and exploitation in terms of the search in the level of deepness that the knowledge is re-used as well as in terms of the scope of the search, place where the knowledge is searched (near or far). The level of deepness is defined as the level that the search for knowledge reviews the knowledge already existent in the firm, i.e., a characteristic of exploitation of knowledge. The scope of the search is the level that new knowledge is explored and it is what characterizes the exploration of new knowledge. The firms can vary the levels of use and re-use of existing knowledge, as well as they can vary the exploration of new knowledge. Ahuja e Katila (2002) argue that the idea of exploitation can be conceptualized in different levels or deepness of search, since the firms can be differentiated not only in the extension of exploring new resources, but also in the extension that they improve their current resources. While the exploration research has a key role in the creation of knowledge as it leads to completely new solutions, exploitation research has the role of combining existing solutions in the generation of new combinations.

Garcia and Nair (2005), consider the terms exploration as a synonym of the traditional research where the result will be innovations, particularly radical and totally new innovations, where the risk is higher and the return is uncertain. Meanwhile, exploitation is related to the development of activities where the firm focuses on the improvement of existing processes, existing products and on the cost efficiency. Exploitation would be related to the development of incremental innovation, with lower risk and less variation (Popadiuk and Choo, 2006; Popadiuk, 2007).

The strategic alliance as a form to explore and exploit knowledge has been the focus of some researchers. Im and Rai (2008) have examined the logistic industry in the United States to determine how the knowledge sharing activities in “long-term interorganizational relationships (IORs)” (p. 1281) is related to the relationship performance gains. Their work shows that the sharing is enabled by “ambidextrous management of the relationship” (p. 1281). Holmqvist (2004) used the collaboration between a software company and its partners in new product development projects to understand the effect of experiential learning processes of exploitation and exploration within and between organizations. Rothaermel and Deeds (2004) show that the type of alliance will influence the format and the timing to use the exploration and the exploitation strategies. Table 1 summarizes the characteristics of exploration and exploitation found in the literature.

Table 1: Characteristics of exploration and exploitation of knowledge

Author	Exploration	Exploitation
March (1991)	Research, search, game, risk, flexibility, discovery, radical innovation, experimenting with new alternatives with uncertainty returns, search for new technology.	Refining, choice, production, efficiency, selection, implementation, execution, extension of current competencies, technologies; with positive, predictable returns, incremental innovation.
Gilsing (2002)	Few choices for coordination mechanisms, such as social procedures and reputation; large variety of knowledge base. Learning and knowledge adjusted accordingly to the emerging environment.	Stability when the dominant design is established. Learning focus is on what is already known. Intensive competition, formal procedures and economy of scale
Li et al. (2006)	Development of new knowledge through the search of knowledge remote in time.	Development of new knowledge through the search of recent knowledge.
Karlson (2005)	Tacit knowledge, with the assumption that tacit knowledge is necessary to develop and acquire new knowledge.	Explicit knowledge that can be applied to refine or improve existing knowledge.
Masini et al. (2004)	A strategy that firm will choose if it is threaten by the possible copy or imitation of its products. High cost of development.		Firms that had chosen to follow a cost leadership strategy, not worried about being copied. Low cost of production.
Prietula; Weingart (2005)	Speculation and uncertainty in terms of defining, refining and alignment of a common knowledge base.		The common knowledge base is already established.

Granger causality method

Granger causality

From Wikipedia, the free encyclopedia

When time Series X Granger causes time series Y, the patterns in X are approximately repeated in Y after some time lag (two examples are indicated with arrows). Thus, past values of X can be used for the prediction of future values ofY.

The Granger causality test is a statistical hypothesis test for determining whether one time series is useful in forecasting another, first proposed in 1969.^[1] Ordinarily,regressions reflect "mere" correlations, but Clive Granger argued that causality in economics could be tested for by measuring the ability to predict the future values of a time series using prior values of another time series. Since the question of "true causality" is deeply philosophical, and because of the post hoc ergo propter hoc fallacy of assuming that one thing preceding another can be used as a proof of causation, econometricians assert that the Granger test finds only "predictive causality".^[2]

A time series X is said to Granger-cause Y if it can be shown, usually through a series of t-tests and F-tests on lagged values of X (and with lagged values of Y also included), that those X values provide statistically significant information about future values of Y.

Granger also stressed that some studies using "Granger causality" testing in areas outside economics reached "ridiculous" conclusions. "Of course, many ridiculous papers appeared", he said in his Nobel Lecture, December 8, 2003.^[3] However, it remains a popular method for causality analysis in time series due to its computational simplicity.^[4][5] The original definition of Granger causality does not account for latent confounding effects and does not capture instantaneous and non-linear causal relationships, though several extensions have been proposed to address these issues.^[4]

Contents

  [hide] 

• 1 Intuition
• 2 Underlying principles
• 3 Method
  • 3.1 Mathematical statement
  • 3.2 Multivariate analysis
• 4 Limitations
• 5 Extensions
• 6 Granger causality in neuroscience
  • 6.1 Extensions to point process models
• 7 See instead
• 8 References
• 9 Further reading

Intuition[edit]

We say that a variable X that evolves over time Granger causes another evolving variable Y if predictions of the value of Y based on its own past values and on the past values of X are better than predictions of Y based only on its own past values.

Underlying principles[edit]

Granger defined the causality relationship based on two principles:^[4][6]

1. The cause happens prior to its effect.
2. The cause has unique information about the future values of its effect.

Given these two assumptions about causality, Granger proposed to test the following hypothesis for identification of a causal effect of  on :

where  refers to probability,  is an arbitrary non-empty set, and  and  respectively denote the information available as of time  in the entire universe, and that in the modified universe in which  is excluded. If the above hypothesis is accepted, we say that  Granger causes .^[4][6]

Method[edit]

If a time series is a stationary process, the test is performed using the level values of two (or more) variables. If the variables are non-stationary, then the test is done using first (or higher) differences. The number of lags to be included is usually chosen using an information criterion, such as the Akaike information criterion or the Schwarz information criterion. Any particular lagged value of one of the variables is retained in the regression if (1) it is significant according to a t-test, and (2) it and the other lagged values of the variable jointly add explanatory power to the model according to an F-test. Then the null hypothesis of no Granger causality is not rejected if and only if no lagged values of an explanatory variable have been retained in the regression.

In practice it may be found that neither variable Granger-causes the other, or that each of the two variables Granger-causes the other.

Mathematical statement[edit]

Let y and x be stationary time series. To test the null hypothesis that x does not Granger-cause y, one first finds the proper lagged values of y to include in a univariate autoregression of y:

Next, the autoregression is augmented by including lagged values of x:

One retains in this regression all lagged values of x that are individually significant according to their t-statistics, provided that collectively they add explanatory power to the regression according to an F-test (whose null hypothesis is no explanatory power jointly added by the x's). In the notation of the above augmented regression, p is the shortest, and q is the longest, lag length for which the lagged value of x is significant.

The null hypothesis that x does not Granger-cause y is not rejected if and only if no lagged values of x are retained in the regression.

Multivariate analysis[edit]

Multivariate Granger causality analysis is usually performed by fitting a vector autoregressive model (VAR) to the time series. In particular, let  for  be a -dimensional multivariate time series. Granger causality is performed by fitting a VAR model with  time lags as follows:

where  is a white Gaussian random vector. A time series  is called a Granger cause of another time series , if at least one of the elements  for  is significantly larger than zero (in absolute value).^[7]

Limitations[edit]

As its name implies, Granger causality is not necessarily true causality. If both X and Y are driven by a common third process with different lags, one might still fail to reject the alternative hypothesis of Granger causality. Yet, manipulation of one of the variables would not change the other. Indeed, the Granger test is designed to handle pairs of variables, and may produce misleading results when the true relationship involves three or more variables. A similar test involving more variables can be applied with vector autoregression.

Extensions[edit]

A method for Granger causality has been developed that is not sensitive to deviations from the assumption that the error term is normally distributed.^[8] This method is especially useful in financial economics since many financial variables are non-normally distributed.^[9] Recently, asymmetric causality testing has been suggested in the literature in order to separate the causal impact of positive changes from the negative ones.^[10]

Granger causality in neuroscience[edit]

A long held belief about neural function maintained that different areas of the brain were task specific; that the structural connectivity local to a certain area somehow dictated the function of that piece. Collecting work that has been performed over many years, there has been a move to a different, network-centric approach to describing information flow in the brain. Explanation of function is beginning to include the concept of networks existing at different levels and throughout different locations in the brain.^[11] The behavior of these networks can be described by non-deterministic processes that are evolving through time. That is to say that given the same input stimulus, you will not get the same output from the network. The dynamics of these networks are governed by probabilities so we treat them as stochastic (random) processes so that we can capture these kinds of dynamics between different areas of the brain.

Different methods of obtaining some measure of information flow from the firing activities of a neuron and its surrounding ensemble have been explored in the past, but they are limited in the kinds of conclusions that can be drawn and provide little insight into the directional flow of information, its effect size, and how it can change with time.^[12] Recently Granger causality has been applied to address some of these issues with great success.^[13] Put plainly, one examines how to best predict the future of a neuron: using either the entire ensemble or the entire ensemble except a certain target neuron. If the prediction is made worse by excluding the target neuron, then we say it has a “g-causal” relationship with the current neuron.

Extensions to point process models[edit]

Previous Granger-causality methods could only operate on continuous-valued data so the analysis of neural spike train recordings involved transformations that ultimately altered the stochastic properties of the data, indirectly altering the validity of the conclusions that could be drawn from it. Recently however, a new general-purpose Granger-causality framework was proposed that could directly operate on any modality, including neural-spike trains.^[12]

Neural spike train data can be modeled as a point-process. A temporal point process is a stochastic time-series of binary events that occurs in continuous time. It can only take on two values at each point in time, indicating whether or not an event has actually occurred. This type of binary-valued representation of information suits the activity of neural populations because a single neuron’s action potential has a typical waveform. In this way, what carries the actual information being output from a neuron is the occurrence of a “spike”, as well as the time between successive spikes. Using this approach one could abstract the flow of information in a neural-network to be simply the spiking times for each neuron through an observation period. A point-process can be represented either by the timing of the spikes themselves, the waiting times between spikes, using a counting process, or, if time is discretized enough to ensure that in each window only one event has the possibility of occurring, that is to say one time bin can only contain one event, as a set of 1s and 0s, very similar to binary.

One of the simplest types of neural-spiking models is the Poisson process. This however, is limited in that it is memory-less. It does not account for any spiking history when calculating the current probability of firing. Neurons, however, exhibit a fundamental (biophysical) history dependence by way of its relative and absolute refractory periods. To address this, a conditional intensity function is used to represent the probability of a neuron spiking,conditioned on its own history. The conditional intensity function expresses the instantaneous firing probability and implicitly defines a complete probability model for the point process. It defines a probability per unit time. So if this unit time is taken small enough to ensure that only one spike could occur in that time window, then our conditional intensity function completely specifies the probability that a given neuron will fire in a certain time.

See instead[edit]

• Transfer entropy
• Convergent cross mapping, another technique for testing for causality between dynamic variables