Description
Most early macroeconomic models, including early Keynesian models, were based on hypotheses about relationships between aggregate quantities, such as aggregate output, employment, consumption, and investment.
How Micro-foundational Thinking Critical to Management Thought and Practice?
This issue of the Academy of Management Perspectives includes a symposium a compelling and controversial set of logics being debated within the management discipline: the microfoundations of management. In this symposium, five papers provide different perspectives on the meaning of microfoundational thinking and its importance to management scholarship. The discussions in these articles—and, indeed, in the literature on microfoundations in general—highlight a number of recurring and yet unresolved issues. The most critical of these is the basic question of what microfoundations research actually entails. Is it dominantly a definitional, or ontological, discussion about what key core concepts and phenomena are of importance to scholars? A second philosophical issue is the extent to which an individualist versus holistic scientific orientation is most appropriate and whether microfoundations is really little more than an ultimately meaningless reductionist exercise. At the more skeptical extreme is Winter (2013). For example, he compares the microfoundations paradigm to work in economics over the past three decades that has been distinctly unsuccessful at linking microeconomic foundations to macroeconomic theory. He also believes that microfoundations can be linked to "long-familiar arguments about reductionism in scientific explanation generally" and classical discussions of individualism versus holism in the social sciences. While Winter goes on to discuss how capabilities and routines can be built up from more fundamental foundations, it is this irreducibility argument and the role of individualist theories
versus holistic theories that we need to bear in mind when asking questions about the appropriateness of microfoundations for management scholarship and thought. As Kincaid (1996) said about irreducibility: [W]holes are, of course, composed of or exhausted by their parts and do not act independently of them; [?] nonetheless, theories at the level of the whole can be confirmed and can explain at that level, without a full accounting of the underlying details; [?] theories at the level of the whole may have only a messy relationship to how micro-level theories divide up the world, thus making macro-level theories irreducible. (p. 142) Kincaid went on to note, not dissimilarly to Winter, that "searching for lower-level accounts can be informative as a complement to, but not a substitute for, more macro investigations and that reduction is not the only route to the ideal of a unified science" (p. 142). At the other end of the spectrum are Barney and Felin (2013), who attempt to bring clarity into the discussion by arguing for what they believe fits into the microfoundations paradigm and outlining an appropriate research agenda. For the most part, they contend that what at times passes for microfoundation scholarship is not truly microfoundational, as the sine qua non of true microfoundations scholarship is aggregation. In other words, "analysis should be fundamentally concerned with how individual-level factors aggregate to the collective level." In making this point they work to counter the "half-truths" of the microfoundation skeptics by arguing that their conception of microfoundations is a distinctive theoretical paradigm and not simply an eclectic collection of micro-level theories or a narrow attempt at reductionism.
Taking up the middle ground are the three papers by Greve (2013), Foss and Lindenberg (2013), and Van de Ven and Lifschitz (2013). All three papers are effectively a search for the right level at which microfoundations research needs to be embedded in our thinking. While they all end up at slightly different points, they are all in agreement that microfoundations is a behavioral paradigm. For Greve, the most appropriate microfoundational level of analysis is at the behavioral strategy level, which for most other scholars in the symposium would no doubt not be micro enough to be considered microfoundational. For Foss and Lindenberg, microfoundations are, at their most basic, psychological, but they work to integrate the aggregation and integration of the actions of goal-seeking economic agents. Van de Ven and Lifschitz take a much less purely psychological orientation to the behavioral issues that arise in the other papers and argue that a "reasonable-person" logic may be a more appropriate basis of microfoundations than one based on a "rational" economic person or a cognitively limited and biased behavioral decision maker. In publishing this symposium, we are concerned not only about the microfoundations paradigm as a distinctive program of scholarship à la Lakatos (1970)—and one with differing perspectives, as evidenced by the papers in this issue—but also in what we see as its role (a) in reflecting on an emerging and growing area of important scholarship and (b) as a trigger for more substantive and nuanced thinking about our work in a more historical and generalized perspective. Let us take each in turn. If we look at microfoundations very simply, we can characterize them as follows (see Figure 1): At the first level we have theories of the individual (what I call I-theories). These are very much theories relating to individual action without recourse to discussions relating to higher levels of analysis. At the second level we
have theories of the collective or the organization (what I call O-theories). These are theories of group interaction and include aspects relating to joint or collective action, routines, and so on. Finally, we have theories at the strategic firm level (what I call Stheories). These are theories relating to firm competition and strategic action and reaction. Note that these are simply characterizations of convenience that no doubt other scholars would find simplistic. However, they are meant only to represent levels of analysis, and scholars can increase or decrease the number of levels should they choose to do so. However, looking at I-, O-, and S-theories individually, microfoundational thinking immediately has value in that we can now begin formally asking very interesting and important scientific questions about what goes on both between and across the levels of analysis. For example, empirically validated I-theories may place emphasis on the risk-averse and biased nature of decision making, while S-theories may operate quite effectively by ignoring that bias and assuming that firms are expected utility maximizers. Similarly, S-theories invariably discuss decision making at the level of the firm, but in most cases it is not the firm actually making the decision but rather an individual manager or a group of managers about whom the researcher invariably has either no information or very little information. In other words, most S-theories are not theories in which the managers matter at all, while most I-theories are not theories in which the firm matters much at all (see, e.g., Devinney, 2011). It is only in the case of O-theories where there is some meshing of firm and individual levels of influence, but exactly how this works has generally been underspecified in the extant research in management. However, this discussion should reveal an intellectual and scientific opportunity that has immense importance to many of our fields of study. We can see this by just
posing a few questions about the implications of different findings across the levels of analysis. For example, why is it that individual managers can be cognitively limited, subject to all sorts of biases, and risk averse while firms do not appear to show the level of bias nor are they as constrained in their decision making in the same way? In the knowledge management field we still cannot answer a very basic question: Is it possible for individuals to be more knowledgeable individually than the firm is collectively? Or alternatively, can the firm possess knowledge that is greater than the sum of the knowledge possessed by its employees? And if so, where does this actually reside? In some of my own work I have questioned whether it is possible for firms to be socially responsible when the individual behaviors of individual stakeholders in and related to the firm—customers, workers, managers, investors—do not show a proclivity to act in a socially responsible manner (see, e.g., Devinney, Auger, & Eckhardt, 2010). Each of these questions is essentially addressing a lack of coherence among the implications of theories and between empirical findings at one level and those at another level. This discussion is critically important for two reasons. First, this lack of coherence implies that we may be missing a critical causal link in the business and management process chain. When we see conflicting findings across levels of analysis, either one or the other theory is lacking in some important detail or there is something related to the aggregation across the levels that is critical in determining the differences in the theoretical structures and empirical findings. This lack of coherence should encourage us to question why these differences exist and whether or not they exist because of something we need to account for theoretically. Microfoundations is, in a sense, a call to action on this generally ignored dimension of our scholarship.
Second, from a practical perspective, predictions and theories that cannot ultimately coalesce across these levels of analysis have limited normative value for a key constituency of our research: managers. Ultimately, managers are concerned with how lower-level actions and activities—what we discuss in our I- and O-theories— affect the performance of the organization. I- and O-theories that cannot be linked to S-theories (which is where performance is located in my simple characterization) may be, as Kincaid said, "informative as a complement to, but not a substitute for, more macro" level theories. Thinking microfoundationally, particularly for those of us who are I-level or O-level researchers, forces us to ask some very basic questions about the value of our own work to those who might use our theories to make actual decisions. Hence, microfoundational thinking can help us do better and more practically important research by forcing us to look across levels of analysis. But this is not as simple as it seems, and hence, goes far beyond just arguing for multilevel analysis. For example, Kincaid (1996) argued that: To reduce one theory to another, we must first have some way to tie the two together, for they describe the world in very different vocabularies. So the first condition [for reduction to work] is that we have some way to translate a holistic theory into the individualistic one by substituting the basic descriptive terms of the individualistic theory for those of the holistic theory. (p. 146) To achieve this requires bridges to be built across the levels of analysis. In this sense Barney and Felin's argument that microfoundations is about the "how" of aggregation is potentially part of the explanation, but it fails as a singular theory of microfoundations. Figure 2 presents a stylized set of such bridges or aggregations. Each higher level of analysis is made up of a functional aggregation of lower-level concepts. I call
these A-theories, or aggregation theories. There are, however, a number of key issues with these theories. First, it is important to note the many-to-many relationships in the examples. In other words, a higher-level concept cannot simply be a lower-level concept rephrased. In other words one cannot simply have I1? O1; additional information must be meaningfully added for a new higher-level concept to arise. Second, just because we can create a stylized example that reveals a clean relationship between lower-level and higher-level concepts does not guarantee that this will arise. It is entirely conceivable that what are known as "multiple realizations" can arise (Shapiro, 2000)—meaning that many different combinations of I's might lead to the same O's, and many different combinations of O's might lead to the same S's. What does this rather philosophical discussion imply for microfoundations? First, it implies that microfoundations is not just about aggregation, nor is aggregation the only essential element. Microfoundations requires researchers to think about the necessary I-theories, O-theories, and S-theories that make up the management system being investigated as well as the A-theories that connect these conceptions in such a way that higher-level O- and S-theories can be explained using I-level conceptualizations. In addition, work in microfoundations has, until this point, been a predominantly theoretical and ontological exercise. However, its value will ultimately be empirical, as multiple realizability implies that what structures are dominant can be answered only by a substantive and sophisticated empirical program (Shapiro, 2008). However, microfoundations can also serve as a base from which we can look at questions from a different perspective. Rather than looking for the gaps in existing narrow theories or inconsistencies across theories at a single level of analysis, microfoundations motivate us to look at how theories at different levels of analysis
differ in their conclusions and whether or not these theories relate to one another (and how they may relate). In publishing this symposium we believe that microfoundations can be a key platform in moving the management field forward. It opens up the possibilities of bridging the macro-micro divide that pervades management research by serving as a conceptual forum to debate whether or not it is possible for us come up with a more unified and parsimonious characterization of our field.
References Barney, J., & Felin, T. (2013). What are microfoundations? Academy of Management Perspectives, 27(2), (this issue). Devinney, T. M. (2011). Bringing managers' decision models into FDI research. In R. Ramamurti & N. Hashai (Eds.), The future of foreign direct investment and the multinational enterprise (pp. 61-83). Bingley, UK: Emerald Group Publishing Limited. Devinney, T. M., Auger, P., & Eckhardt, G. M. (2010). The myth of the ethical consumer. Cambridge, UK: Cambridge University Press. Foss, N. J., & Lindenberg, S. (2013). Microfoundations for Strategy: A Goal-Framing Perspective on the Drivers of Value Creation. Academy of Management Perspectives, 27(2), (this issue). Greve, H. R. (2013). Microfoundations of Management: Behavioral Strategies and Levels of Rationality in Organizational Action. Academy of Management Perspectives, 27(2), (this issue). Kincaid, H. (1996). Philosophical foundations of the social sciences: Analyzing controversies in social research. Cambridge, UK: Cambridge University Press. Lakatos, I. (1970). Falsification and the methodology of scientific research programmes. In I. Lakatos & A. Musgrave (Eds.), Criticism and the growth of knowledge (pp. 170-196). Cambridge, UK: Cambridge University Press. Shapiro, L. (2000). Multiple realizations. Journal of Philosophy, 97(12), 635-654. Shapiro, L. (2008). How to test for multiple realization. Philosophy of Science, 75(5), 514-525. Van de Ven, A. H., and Lifschitz, A. (2013). Rational and Reasonable Microfoundations of Markets and Institutions. Academy of Management Perspectives, 27(2), (this issue).
Winter, S. G. (2013). Habit, Deliberation, and Action: Strengthening the Microfoundations of Routines and Capabilities. Academy of Management Perspectives, 27(2), (this issue).
FIGURE 1. I-Level, O-Level, and S-Level Theories
FIGURE 2. Microfoundations as I-Level, O-Level, and S-Level Theories Bridged
doc_505150709.docx
Most early macroeconomic models, including early Keynesian models, were based on hypotheses about relationships between aggregate quantities, such as aggregate output, employment, consumption, and investment.
How Micro-foundational Thinking Critical to Management Thought and Practice?
This issue of the Academy of Management Perspectives includes a symposium a compelling and controversial set of logics being debated within the management discipline: the microfoundations of management. In this symposium, five papers provide different perspectives on the meaning of microfoundational thinking and its importance to management scholarship. The discussions in these articles—and, indeed, in the literature on microfoundations in general—highlight a number of recurring and yet unresolved issues. The most critical of these is the basic question of what microfoundations research actually entails. Is it dominantly a definitional, or ontological, discussion about what key core concepts and phenomena are of importance to scholars? A second philosophical issue is the extent to which an individualist versus holistic scientific orientation is most appropriate and whether microfoundations is really little more than an ultimately meaningless reductionist exercise. At the more skeptical extreme is Winter (2013). For example, he compares the microfoundations paradigm to work in economics over the past three decades that has been distinctly unsuccessful at linking microeconomic foundations to macroeconomic theory. He also believes that microfoundations can be linked to "long-familiar arguments about reductionism in scientific explanation generally" and classical discussions of individualism versus holism in the social sciences. While Winter goes on to discuss how capabilities and routines can be built up from more fundamental foundations, it is this irreducibility argument and the role of individualist theories
versus holistic theories that we need to bear in mind when asking questions about the appropriateness of microfoundations for management scholarship and thought. As Kincaid (1996) said about irreducibility: [W]holes are, of course, composed of or exhausted by their parts and do not act independently of them; [?] nonetheless, theories at the level of the whole can be confirmed and can explain at that level, without a full accounting of the underlying details; [?] theories at the level of the whole may have only a messy relationship to how micro-level theories divide up the world, thus making macro-level theories irreducible. (p. 142) Kincaid went on to note, not dissimilarly to Winter, that "searching for lower-level accounts can be informative as a complement to, but not a substitute for, more macro investigations and that reduction is not the only route to the ideal of a unified science" (p. 142). At the other end of the spectrum are Barney and Felin (2013), who attempt to bring clarity into the discussion by arguing for what they believe fits into the microfoundations paradigm and outlining an appropriate research agenda. For the most part, they contend that what at times passes for microfoundation scholarship is not truly microfoundational, as the sine qua non of true microfoundations scholarship is aggregation. In other words, "analysis should be fundamentally concerned with how individual-level factors aggregate to the collective level." In making this point they work to counter the "half-truths" of the microfoundation skeptics by arguing that their conception of microfoundations is a distinctive theoretical paradigm and not simply an eclectic collection of micro-level theories or a narrow attempt at reductionism.
Taking up the middle ground are the three papers by Greve (2013), Foss and Lindenberg (2013), and Van de Ven and Lifschitz (2013). All three papers are effectively a search for the right level at which microfoundations research needs to be embedded in our thinking. While they all end up at slightly different points, they are all in agreement that microfoundations is a behavioral paradigm. For Greve, the most appropriate microfoundational level of analysis is at the behavioral strategy level, which for most other scholars in the symposium would no doubt not be micro enough to be considered microfoundational. For Foss and Lindenberg, microfoundations are, at their most basic, psychological, but they work to integrate the aggregation and integration of the actions of goal-seeking economic agents. Van de Ven and Lifschitz take a much less purely psychological orientation to the behavioral issues that arise in the other papers and argue that a "reasonable-person" logic may be a more appropriate basis of microfoundations than one based on a "rational" economic person or a cognitively limited and biased behavioral decision maker. In publishing this symposium, we are concerned not only about the microfoundations paradigm as a distinctive program of scholarship à la Lakatos (1970)—and one with differing perspectives, as evidenced by the papers in this issue—but also in what we see as its role (a) in reflecting on an emerging and growing area of important scholarship and (b) as a trigger for more substantive and nuanced thinking about our work in a more historical and generalized perspective. Let us take each in turn. If we look at microfoundations very simply, we can characterize them as follows (see Figure 1): At the first level we have theories of the individual (what I call I-theories). These are very much theories relating to individual action without recourse to discussions relating to higher levels of analysis. At the second level we
have theories of the collective or the organization (what I call O-theories). These are theories of group interaction and include aspects relating to joint or collective action, routines, and so on. Finally, we have theories at the strategic firm level (what I call Stheories). These are theories relating to firm competition and strategic action and reaction. Note that these are simply characterizations of convenience that no doubt other scholars would find simplistic. However, they are meant only to represent levels of analysis, and scholars can increase or decrease the number of levels should they choose to do so. However, looking at I-, O-, and S-theories individually, microfoundational thinking immediately has value in that we can now begin formally asking very interesting and important scientific questions about what goes on both between and across the levels of analysis. For example, empirically validated I-theories may place emphasis on the risk-averse and biased nature of decision making, while S-theories may operate quite effectively by ignoring that bias and assuming that firms are expected utility maximizers. Similarly, S-theories invariably discuss decision making at the level of the firm, but in most cases it is not the firm actually making the decision but rather an individual manager or a group of managers about whom the researcher invariably has either no information or very little information. In other words, most S-theories are not theories in which the managers matter at all, while most I-theories are not theories in which the firm matters much at all (see, e.g., Devinney, 2011). It is only in the case of O-theories where there is some meshing of firm and individual levels of influence, but exactly how this works has generally been underspecified in the extant research in management. However, this discussion should reveal an intellectual and scientific opportunity that has immense importance to many of our fields of study. We can see this by just
posing a few questions about the implications of different findings across the levels of analysis. For example, why is it that individual managers can be cognitively limited, subject to all sorts of biases, and risk averse while firms do not appear to show the level of bias nor are they as constrained in their decision making in the same way? In the knowledge management field we still cannot answer a very basic question: Is it possible for individuals to be more knowledgeable individually than the firm is collectively? Or alternatively, can the firm possess knowledge that is greater than the sum of the knowledge possessed by its employees? And if so, where does this actually reside? In some of my own work I have questioned whether it is possible for firms to be socially responsible when the individual behaviors of individual stakeholders in and related to the firm—customers, workers, managers, investors—do not show a proclivity to act in a socially responsible manner (see, e.g., Devinney, Auger, & Eckhardt, 2010). Each of these questions is essentially addressing a lack of coherence among the implications of theories and between empirical findings at one level and those at another level. This discussion is critically important for two reasons. First, this lack of coherence implies that we may be missing a critical causal link in the business and management process chain. When we see conflicting findings across levels of analysis, either one or the other theory is lacking in some important detail or there is something related to the aggregation across the levels that is critical in determining the differences in the theoretical structures and empirical findings. This lack of coherence should encourage us to question why these differences exist and whether or not they exist because of something we need to account for theoretically. Microfoundations is, in a sense, a call to action on this generally ignored dimension of our scholarship.
Second, from a practical perspective, predictions and theories that cannot ultimately coalesce across these levels of analysis have limited normative value for a key constituency of our research: managers. Ultimately, managers are concerned with how lower-level actions and activities—what we discuss in our I- and O-theories— affect the performance of the organization. I- and O-theories that cannot be linked to S-theories (which is where performance is located in my simple characterization) may be, as Kincaid said, "informative as a complement to, but not a substitute for, more macro" level theories. Thinking microfoundationally, particularly for those of us who are I-level or O-level researchers, forces us to ask some very basic questions about the value of our own work to those who might use our theories to make actual decisions. Hence, microfoundational thinking can help us do better and more practically important research by forcing us to look across levels of analysis. But this is not as simple as it seems, and hence, goes far beyond just arguing for multilevel analysis. For example, Kincaid (1996) argued that: To reduce one theory to another, we must first have some way to tie the two together, for they describe the world in very different vocabularies. So the first condition [for reduction to work] is that we have some way to translate a holistic theory into the individualistic one by substituting the basic descriptive terms of the individualistic theory for those of the holistic theory. (p. 146) To achieve this requires bridges to be built across the levels of analysis. In this sense Barney and Felin's argument that microfoundations is about the "how" of aggregation is potentially part of the explanation, but it fails as a singular theory of microfoundations. Figure 2 presents a stylized set of such bridges or aggregations. Each higher level of analysis is made up of a functional aggregation of lower-level concepts. I call
these A-theories, or aggregation theories. There are, however, a number of key issues with these theories. First, it is important to note the many-to-many relationships in the examples. In other words, a higher-level concept cannot simply be a lower-level concept rephrased. In other words one cannot simply have I1? O1; additional information must be meaningfully added for a new higher-level concept to arise. Second, just because we can create a stylized example that reveals a clean relationship between lower-level and higher-level concepts does not guarantee that this will arise. It is entirely conceivable that what are known as "multiple realizations" can arise (Shapiro, 2000)—meaning that many different combinations of I's might lead to the same O's, and many different combinations of O's might lead to the same S's. What does this rather philosophical discussion imply for microfoundations? First, it implies that microfoundations is not just about aggregation, nor is aggregation the only essential element. Microfoundations requires researchers to think about the necessary I-theories, O-theories, and S-theories that make up the management system being investigated as well as the A-theories that connect these conceptions in such a way that higher-level O- and S-theories can be explained using I-level conceptualizations. In addition, work in microfoundations has, until this point, been a predominantly theoretical and ontological exercise. However, its value will ultimately be empirical, as multiple realizability implies that what structures are dominant can be answered only by a substantive and sophisticated empirical program (Shapiro, 2008). However, microfoundations can also serve as a base from which we can look at questions from a different perspective. Rather than looking for the gaps in existing narrow theories or inconsistencies across theories at a single level of analysis, microfoundations motivate us to look at how theories at different levels of analysis
differ in their conclusions and whether or not these theories relate to one another (and how they may relate). In publishing this symposium we believe that microfoundations can be a key platform in moving the management field forward. It opens up the possibilities of bridging the macro-micro divide that pervades management research by serving as a conceptual forum to debate whether or not it is possible for us come up with a more unified and parsimonious characterization of our field.
References Barney, J., & Felin, T. (2013). What are microfoundations? Academy of Management Perspectives, 27(2), (this issue). Devinney, T. M. (2011). Bringing managers' decision models into FDI research. In R. Ramamurti & N. Hashai (Eds.), The future of foreign direct investment and the multinational enterprise (pp. 61-83). Bingley, UK: Emerald Group Publishing Limited. Devinney, T. M., Auger, P., & Eckhardt, G. M. (2010). The myth of the ethical consumer. Cambridge, UK: Cambridge University Press. Foss, N. J., & Lindenberg, S. (2013). Microfoundations for Strategy: A Goal-Framing Perspective on the Drivers of Value Creation. Academy of Management Perspectives, 27(2), (this issue). Greve, H. R. (2013). Microfoundations of Management: Behavioral Strategies and Levels of Rationality in Organizational Action. Academy of Management Perspectives, 27(2), (this issue). Kincaid, H. (1996). Philosophical foundations of the social sciences: Analyzing controversies in social research. Cambridge, UK: Cambridge University Press. Lakatos, I. (1970). Falsification and the methodology of scientific research programmes. In I. Lakatos & A. Musgrave (Eds.), Criticism and the growth of knowledge (pp. 170-196). Cambridge, UK: Cambridge University Press. Shapiro, L. (2000). Multiple realizations. Journal of Philosophy, 97(12), 635-654. Shapiro, L. (2008). How to test for multiple realization. Philosophy of Science, 75(5), 514-525. Van de Ven, A. H., and Lifschitz, A. (2013). Rational and Reasonable Microfoundations of Markets and Institutions. Academy of Management Perspectives, 27(2), (this issue).
Winter, S. G. (2013). Habit, Deliberation, and Action: Strengthening the Microfoundations of Routines and Capabilities. Academy of Management Perspectives, 27(2), (this issue).
FIGURE 1. I-Level, O-Level, and S-Level Theories
FIGURE 2. Microfoundations as I-Level, O-Level, and S-Level Theories Bridged
doc_505150709.docx