Academy of Management Annals

Transaction Cost Theory: Past Progress, Current Challenges,

Journal: Academy of Management Annals

Manuscript ID ANNALS-2019-0051.R4

Document Type: Article

GOVERNANCE AND CONTROL, management < INTERNATIONAL,

1
2
3 TRANSACTION COST THEORY: PAST PROGRESS, CURRENT CHALLENGES,
4
5 AND SUGGESTIONS FOR THE FUTURE
6
7
8
9
Ilya R. P. Cuypers
10 Lee Kong Chian School of Business
11
12 Singapore Management University
13 [email protected]
14
15
16
17 Jean-François Hennart
18 Department of Management, Tilburg School of Economics and Management
19
20 Tilburg University
21
22
Department of Management, Economics and Industrial Engineering,
23 Politecnico di Milano
24
25 International Business Centre, Aalborg University Business School
26 Aalborg University
27
28 [email protected]
29
30
31 Brian S. Silverman
32
33 Rotman School of Management
34
35
University of Toronto
36 [email protected]
37
38
39
Gokhan Ertug
40
41 Lee Kong Chian School of Business
42
43 Singapore Management University
44 [email protected]
45
46
47
48
49
50
51 Acknowledgments: We would like to thank our reviewers for their helpful comments. We are
52
53
grateful for the excellent editorial guidance and direction of Bill McEvily. We would also like to
54 thank Jackson Nickerson and Joanne Oxley for valuable conversations that improved this work.
55 Standard disclaimers apply.
56
57
58 1
59
60
 Academy of Management Annals Page 2 of 90

1
2
3
4
5
6
7
8 TRANSACTION COST THEORY: PAST PROGRESS, CURRENT CHALLENGES,
9
10 AND SUGGESTIONS FOR THE FUTURE
11
12
13
14
15
16
17
18
19
Abstract
20
21 Transaction cost theory (TCT) has been fruitfully applied to a wide range of
22
organizational phenomena, as reflected in a vast and evolving body of research.
23
24 However, in part due to the theory’s broad success, important advances in some fields
25 do not diffuse to other fields. In this essay, we lay out a path toward a pluralistic view
26 of TCT that incorporates insights from multiple fields, primarily strategy and
27
28
international business. In so doing, we critically assess the assumptions, key
29 constructs, and evolving theoretical logic of TCT. We then propose an agenda for
30 future research, highlighting opportunities for scholars to (1) expand and deepen the
31 exchange of insights between strategy and international business, and further
32
33
integrate TCT with the trust literature and with recent insights from behavioral
34 economics and psychology, and (2) further apply TCT to study recent phenomena
35 such as platforms and two-sided markets, the implications of artificial intelligence for
36 governance decisions, and the pursuit of non-pecuniary objectives such as
37
38 sustainability.
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 2
59
60
Page 3 of 90 Academy of Management Annals

1
2
3 INTRODUCTION 1
4
5
6 Building on the seminal work of two Nobel laureates (Coase, 1937; Williamson, 1975,
7
8 1985), transaction cost theory (“TCT”), or transaction cost economics, has become one of the most
9
10 influential theories in management research. Originally applied to the “make-vs.-buy” vertical
11
12
13
integration decision, TCT has been used to shed light on a broad range of organizational
14
15 phenomena, including horizontal diversification, the multinational enterprise, strategic alliances,
16
17 supply chain relationships, and public-private partnerships. TCT has also expanded to encompass
18
19
an increasing roster of factors that predict governance choice, as well as the performance
20
21
22 consequences of this choice. This expansion is evidence of the theory’s influence and success.
23
24 However, that influence and success is a double-edged sword. Given the breadth of its application,
25
26 the TCT literature risks becoming fragmented and difficult to navigate, as advances in one sphere
27
28
29 are overlooked by others, and as definitions of key concepts evolve differently across these
30
31 spheres.
32
33 The purpose of this article is to provide a roadmap for navigating this vast literature and to
34
35
36
offer suggestions for conceptual and empirical future work on TCT. Whereas prior surveys of TCT
37
38 have summarized the broad empirical state of the field (e.g., David & Han, 2004) or focused on a
39
40 empirical phenomena such as vertical integration and entry mode choices (e.g., Geyskens,
41
42
Steenkamp, & Kumar, 2006; Zhao, Luo, & Suh, 2004), our survey emphasizes the distinctive
43
44
45 features of TCT research across different conceptual areas. We first provide an overview of the
46
47 basic proposition of TCT as initially formulated by Williamson (e.g., Williamson, 1973, 1975),
48
49 revisiting its underlying assumptions, key theoretical constructs, and logic. We then illustrate how
50
51
52
53 1While this article was under review, Oliver Williamson passed away. This is a particularly bittersweet time to reflect
54 upon the theory with which he is so closely associated, and to affirm that one of the joyous features of academe is that
55 foundational ideas may be sparked by one or a few individuals, but then evolve to have a life of their own as they are
56 extended and refined by the academic community.
57
58 3
59
60
 Academy of Management Annals Page 4 of 90

1
2
3 the literature has evolved from this initial formulation, paying particular attention to contributions
4
5
6 from the strategy and international business literatures. Of particular note is that whereas TCT
7
8 initially emphasized asset specificity as the most important of the three attributes affecting
9
10 governance, subsequent advances have highlighted the importance of behavioral uncertainty, and
11
12
13
have also considered factors beyond that initial set of three attributes. To spur future research in
14
15 TCT, we conclude with a call for research that can apply these insights to three conceptual areas
16
17 – further integration of research in strategy, international business, and institutional economics;
18
19
greater engagement with research in sociology, in particular the literature on trust and the research
20
21
22 on formal-informal organization; and greater links to the psychology and behavioral economics
23
24 literatures for a better understanding of the behavioral processes that frame the governance of
25
26 transactions. In addition, we highlight three phenomenological areas for future research – platform
27
28
29 governance, technological advances such as artificial intelligence and machine learning, and the
30
31 increased prevalence of non-pecuniary goals for a range of actors (e.g., the rise of nationalism,
32
33 corporate social responsibility, and “grand challenges”). Given these conceptual and empirical
34
35
36
opportunities for further development and application, we are confident that TCT’s future is as
37
38 bright as its past.
39
40 THE BASIC PROPOSITION
41
42
TCT’s fundamental prediction is that organizational actors attempt to maximize the gains
43
44
45 of interdependence by “assigning transactions (which differ in their attributes) to governance
46
47 structures (the adaptive capacities and associated costs of which differ) in a discriminating way”
48
49 (Williamson, 1985: 18). Building on the behavioral assumptions of bounded rationality and
50
51
52 opportunism, Williamson’s TCT argues that transactions will be assigned to governance structures
53
54 based on three key attributes – asset specificity, uncertainty, and frequency. According to
55
56
57
58 4
59
60
Page 5 of 90 Academy of Management Annals

1
2
3 Williamson, the most important of these three attributes is asset specificity, which is the degree of
4
5
6 specific investment involved in a transaction. An asset is specific to a particular transaction if its
7
8 value in its next-best use (i.e., in a transaction with a different party) is lower than in the present
9
10 transaction. The greater the difference between the value of an asset in its first-best and next-best
11
12
13
use, the greater the degree of asset specificity (Klein, Crawford, & Alchian, 1978; Williamson,
14
15 1979).
16
17 When exchange hazards are not significant or are negligible – at the limit, when the assets
18
19
that underpin and facilitate a transaction are generic – spot markets offer the least-costly form of
20
21
22 governance. Markets provide strong incentives for effort and for autonomous adaptation, and
23
24 parties incur few or no private set-up costs. Since the exchange relies on generic assets, disputes
25
26 that might arise between transacting parties can be settled at low cost by exiting the exchange. In
27
28
29 contrast, when assets are transaction-specific, hierarchy is the least-costly governance solution.
30
31 Although hierarchy entails high private fixed set-up costs and reduces incentives to maximize
32
33 outputs, it facilitates a coordination of investments and activities that is challenging to manage
34
35
36
through markets. Within a hierarchy, authority (“fiat”) can ultimately settle disagreements about
37
38 the nature and allocation of tasks. These different governance arrangements are also supported by
39
40 different legal regimes, which range from court enforced contract law for some market transactions
41
42
to internal enforcement for hierarchy (Masten, 1988). In equilibrium, organizational actors are
43
44
45 predicted to choose the appropriate organizational form to govern a given transaction. We
46
47 summarize the basic proposition of Williamson’s TCT framework and its key components in
48
49 Figure 1.
50
51
52 <<< INSERT FIGURE 1 HERE >>>
53
54
55
56
57
58 5
59
60
 Academy of Management Annals Page 6 of 90

1
2
3 Over the years, this basic proposition has evolved in a number of ways, driven both by the
4
5
6 application of TCT ideas to new types of transactions and by broad changes to modern economic
7
8 activity. International business scholars, who had independently begun work on firm-boundary
9
10 issues (Buckley & Casson, 1976; Rugman, 1981), were early movers in integrating and extending
11
12
13
these ideas to inform scholarship on international expansion and the structure of the multinational
14
15 enterprise (Hennart, 1982). Perhaps not surprisingly, given the importance of transferring
16
17 knowledge (rather than specialized physical assets) across geographic locations, issues of
18
19
behavioral uncertainty became particularly prominent in this literature. Related to this, and to deal
20
21
22 with the growth of the knowledge-based economy, strategy scholars introduced appropriability -
23
24 which refers to the degree to which an economic actor can protect its knowledge from leakage to
25
26 other parties (Teece, 1986) - as another important factor that influences governance choices (e.g.,
27
28
29 Teece, 1986; Oxley, 1997; Silverman, 1999). As the business world experienced an explosion in
30
31 collaborative activity, TCT sharply increased its focus on governance modes that incorporate
32
33 aspects of both market and hierarchy. This has led to many insights, but also resulted in the
34
35
36
generation of a number of alternative views. For example, some view alliances and other “hybrids”
37
38 as a distinct governance mode (Williamson, 1991). Others, however, distinguish between
39
40 governance mechanisms (the price system and hierarchy) and institutions (markets, firms, and
41
42
hybrids), and see alliances and hybrids as institutions, rather than as a distinct governance mode,
43
44
45 that use different mixes of prices and hierarchy (Hennart, 1993, 2013). We will further discuss
46
47 these advances and alternative views later in our review.
48
49 THE BEHAVIORAL ASSUMPTIONS
50
51
52 Bounded rationality
53
54
55
56
57
58 6
59
60
Page 7 of 90 Academy of Management Annals

1
2
3 The first assumption underlying TCT is that of bounded rationality. This assumption is
4
5
6 broadly related to that of many perspectives rooted in economics – namely, that there are costs to
7
8 the collection and analysis of information, and consequently different actors will have access to
9
10 different sets of information (e.g., Holmstrom & Milgrom, 1991; Arrow, 1974). TCT’s assumption
11
12
13
of bounded rationality goes beyond information costs to recognize that agents have limits to their
14
15 analytical and data-processing abilities, and that therefore these agents experience constraints in
16
17 processing information and in formulating and solving complex problems even when information
18
19
is available.2 Put simply, economic actors are “intendedly rational, but only limitedly so” (Simon,
20
21
22 1957: xxiv; Williamson, 1981: 553). This concept of bounded rationality is distinct from both
23
24 irrationality and hyperrationality. Of particular relevance to the theory is the idea that, if rationality
25
26 were not bounded – that is, if economic actors could costlessly anticipate every future contingency
27
28
29 – then they could write complete contracts covering any potential outcome. In this situation, there
30
31 would be no transaction costs associated with contracting. Consequently, all transactions could be
32
33 effected through the market (Williamson, 1981).3
34
35
36
Opportunism
37
38 The second assumption underlying TCT is that of opportunism, which Williamson (1975:
39
40 255) famously defined as “self-interest seeking with guile.” In other words, actors do not always
41
42
share full information, provide objective assessments of likely outcomes, or behave cooperatively
43
44
45 during the execution of economic exchanges. As with bounded rationality, opportunism is
46
47 consistent with many economists’ view of self-interested behavior, in which actors take actions
48
49
50 2 If the only thing that matters is the cost of information, then mechanisms that lead to information revelation can
51
solve all problems. But if there are also limits to people’s ability to process and analyze information, then such
52
mechanisms alone will not be sufficient.
53 3 Absent bounded rationality, the cost of organizing activities in a hierarchical manner might also become negligible.
54 Hence, in such a scenario, both hierarchy and the price system could efficiently organize interdependencies, making
55 the choice of governance irrelevant (Hennart, 1982).
56
57
58 7
59
60
 Academy of Management Annals Page 8 of 90

1
2
3 that maximize their payoffs (e.g., Grossman & Hart, 1986; Hart & Moore, 1990; Holmstrom &
4
5
6 Milgrom, 1991), but goes beyond this to recognize that actors may also make acts of omission or
7
8 commission that tilt payoffs in their favor. This has a pervasive impact on economic organization.
9
10 In a world without any risk of opportunism, actors could always contract through the market and
11
12
13
simply agree to “work things out” as future events unfold. Governance would play no role in a
14
15 world in which all people will honor their promises all the time. But in a world with opportunism
16
17 (coupled with bounded rationality), market exchange can be fraught with hazards.
18
19
The assumption of opportunism has been more controversial than that of bounded
20
21
22 rationality (Williamson, 1993), and has sometimes been construed to imply that most or all people
23
24 are prepared to lie, cheat, and steal most or all of the time (e.g., Perrow, 1986). Several prominent
25
26 scholars have argued that this assumption is actually damaging to the business world, because it
27
28
29 encourages managers to assume the worst about exchange partners or, worse, seek to be pre-
30
31 emptively opportunistic themselves (Ghoshal & Moran, 1996; Ghoshal, 2005; Pfeffer, 2005). We
32
33 note that, despite Williamson’s occasional use of a florid phrase (oftentimes a quotation from
34
35
36
another social scientist), TCT does not propose that all actors behave opportunistically, but rather
37
38 that some small proportion of actors will behave opportunistically some small proportion of the
39
40 time, and that it is difficult to predict who will be opportunistic, and when. This small possibility
41
42
of opportunistic behavior can be enough to cause market exchange to break down.4 Further, TCT
43
44
45
46
4 Prior work in economics had highlighted the problems of moral hazard and adverse selection with specific respect
47
48 to insurance (e.g., Arrow, 1971), yet the field had not yet grappled with the underlying issue – that people might not
49 always truthfully reveal their “type” (i.e., health) or honor pledges (i.e., to exercise or forgo smoking) – in other
contexts. Williamson (1975: 5) is explicit that “the insurance problem [..] is really a paradigm for studying the
50
employment relation, vertical integration, and competition in the capital market.” He defines opportunism thus:
51
“Opportunism is an effort to realize individual gains through a lack of candor or honesty in transactions. It can take
52
either of two forms. The most commonly recognized is the strategic disclosure of asymmetrically distributed
53 information by (at least some) individuals to their advantage. Original negotiations may be impaired on this account.
54 The second type manifests itself during contract execution and renewal. The impossibility of extracting what can be
55 confidently regarded as self-enforcing promises to behave ‘responsibly’ requires that agreements be monitored and
56 may pose problems, due to first-mover advantages, at the contract renewal interval [..]” (Williamson, 1973: 317). This
57
58 8
59
60
Page 9 of 90 Academy of Management Annals

1
2
3 scholars do not celebrate the existence of opportunism. As Ketokivi and Mahoney (2016: 129,
4
5
6 footnote 5) point out, management scholars “spend a lot of time thinking about [and researching]
7
8 insurable risks…. Does this mean that [they] condone fires and other calamities? Of course it does
9
10 not.” In analogous fashion, TCT scholars think about transaction risks in order to devise safeguards
11
12
13
that will allow economic exchange to occur even in the face of these risks. Simply put, opportunism
14
15 is important to this thought process.
16
17 In the international business literature, Verbeke and Greidanus (2009) have offered a broader
18
19
version of opportunism, which they call bounded reliability. They see it as an envelope concept, with
20
21
22 two main components: (1) the Williamsonian dimension of strategic opportunism, but also (2) non-
23
24 strategic preference reversals, which are ex post reordering of commitments that were made in good
25
26 faith ex ante, due to changes in priorities and overcommitment. The authors assert that most cases
27
28
29 of opportunism are due to the latter rather than to the former. In other words, opportunism arises
30
31 because “most actors are reliable, but only boundedly so” (Kano & Verbeke, 2015: 98) and their
32
33 failure to fulfil their commitment is more often due to reprioritization and overcommitment than to
34
35
36
a deliberate intent to deceive and cheat. Verbeke and Greidanus (2009) argue that bounded reliability
37
38
39
40 is later elaborated as “the strategic manipulation of information and misrepresentation of intentions.” Without the
41 latter of these, “self-enforcing promises to the effect that ‘I solemnly pledge to execute this contract efficiently and to
42 seek only fair returns at the contract renewal interval” would be sufficient to eliminate transaction costs. However, “at
43 least some of the agents who accede to such terms do it casually, in a self-disbelieved way. Since these types cannot
44 be distinguished ex ante from sincere types,” contractual hazards arise (Williamson 1975:26-27). (For a more complete
45 discussion, see Williamson [1993].)
46 In his critique of organizational economics, Perrow (1986: 12) describes opportunism thus: “monitoring contracts
47 is costly and somewhat ineffective, especially in organizations, thus encouraging self-interested behavior, shirking,
48 and especially opportunism with guile, or to put it more simply – cheating. Contracts will be violated because of self-
49 interest, and can be violated because of the costs and ineffectiveness of surveillance [..] [T]ransaction-cost analysis,
assumes that ‘human nature as we know it’, as Oliver Williamson puts it, is prone to opportunism with guile.” Perrow’s
50
description ascribes to Williamson a view about the pervasiveness of opportunism that Williamson does not evince.
51
Ghoshal and Moran (1996: 14) quote Williamson as follows: “[Opportunism] allows for ‘strategic behavior,’ that
52
is, ‘the making of false or empty, that is, self-disbelieved, threats and promises in the expectation that individual
53 advantage will thereby be realized (Williamson, 1975: 26)” (Ghoshal & Moran 1996: 17, emphasis added). This may
54 appear to be a more provocative definition due to some of the word choices in the italicized phrase. It is worth noting
55 that Williamson did not coin this phrase – rather, it is a quotation from Erving Goffman (1969: 88) within Williamson’s
56 sentence, an attribution that did not make it into the Ghoshal and Moran article.
57
58 9
59
60
 Academy of Management Annals Page 10 of 90

1
2
3 is a more accurate behavioral assumption than Williamsonian opportunism, because it does not
4
5
6 require making the assumption that human beings are strategically untrustworthy. Instead, “they
7
8 have a propensity to make imperfect effort to make good on commitments” (Verbeke & Greidanus,
9
10 2009: 1490).5
11
12
13
We believe that further exploration of bounded reliability is a productive direction to
14
15 pursue. As discussed above, the concept is more general than strategic opportunism, which may
16
17 allow it to explain a broader or different set of governance decisions. It may more faithfully reflect
18
19
what happens in practice in some settings - particularly for a transaction that takes place over a
20
21
22 long period where there is a risk that one or both of the parties will change their priorities. This is
23
24 a transactional hazard, but it does not require ex ante premeditation. One key issue for work in this
25
26 area is under which circumstances bounded reliability will lead to predictions that differ from those
27
28
29 of opportunism, and why.
30
31 Revisiting the behavioral assumptions
32
33 Many areas in management research have increasingly embraced a more behavioral focus.
34
35
36
For example, scholars have looked at the micro-foundations of dynamic capabilities by focusing
37
38 on the sensing and seizing of opportunities (Teece, 2007). Considering that the assumptions at the
39
40 foundation of TCT are behavioral in nature, and that bounded rationality, in particular, clearly
41
42
acknowledges limits on the rationality of decision-making, it would appear that TCT is well-
43
44
45 positioned to benefit from this trend. However, relatively few TCT scholars have turned their
46
47 attention to fleshing out the behavioral aspects of the theory. A small body of work has explored
48
49 the role of risk preferences and the perception of behavioral uncertainty (i.e., likelihood of
50
51
52
53
54 5Williamson’s later work (e.g., 1991) increasingly emphasized the challenge of coordinated adaptation in the face of
55 unanticipated events that could differentially influence the private benefits of parties to an exchange, leading to what
56 Verbeke and Greidanus (2009) might call reprioritization.
57
58 10
59
60
Page 11 of 90 Academy of Management Annals

1
2
3 opportunism) in refining the TCT logic (e.g., Chiles & McMackin, 1996; Buckley & Chapman,
4
5
6 1997; Weber & Mayer, 2014). One key insight from this work is that the way in which a contract
7
8 is framed can influence perceptions of cooperativeness. For example, Weber, Mayer, and Macher
9
10 (2011) note that contractual clauses concerning early termination and extendability are essentially
11
12
13
identical as safeguards – a five-year contract that allows for a second five-year term if both parties
14
15 opt in is the same as a ten-year contract that allows for either party to opt out at the five-year mark.
16
17 The authors theorize, however, that these clauses frame the safeguard in distinct ways that can
18
19
create different psychological stances toward the exchange, and they uncover empirical
20
21
22 regularities in the use of each clause that are consistent with the notion that actors use renewal
23
24 clauses when they seek to emphasize the beneficial prospect of future cooperation.
25
26 As we elaborate below, we envision significant potential for scholars to further integrate
27
28
29 insights from decision-making and psychology into TCT. While the consideration and
30
31 incorporation of behavioral factors may not alter TCT’s prescriptions for how managers should
32
33 make decisions, work on heuristics, attention, risk preferences, and biases (e.g., Kahneman et al.,
34
35
36
1982; Ariely, 2008; Gigerenzer & Gaissmaier, 2011; Kahneman, 2011) would improve our
37
38 understanding of how behavioral uncertainty is perceived and incorporated into decision-makers’
39
40 analyses. This could enable us to better explain what kind of decisions managers do make, and
41
42
why.
43
44
45 Separately, there exists some tension in the TCT literature due to differing interpretations
46
47 of the extent to which rationality is bounded. On the one hand, bounded rationality is assumed to
48
49 constrain actors’ ex ante forethought sufficiently for transaction costs to emerge. On the other
50
51
52 hand, it is assumed to be insufficiently constraining to systematically hamper their ability to make
53
54 appropriate governance choices in the face of future unknowns. The TCT literature has
55
56
57
58 11
59
60
 Academy of Management Annals Page 12 of 90

1
2
3 accumulated evidence showing that firms, on average, do indeed “get it right,” in the sense of
4
5
6 making governance decisions that accord with the prescriptions of TCT (e.g., Geyskens et al.,
7
8 2006; Zenger, Felin, & Bigelow, 2011). However, several studies have also highlighted that some
9
10 firms appear to make choices that are suboptimal from a TCT perspective (e.g., Sampson, 2004).
11
12
13
The current TCT literature has not offered systematic explanations for the sources of these
14
15 deviations, yet their presence and patterns might be partially predictable given the bounded
16
17 rationality of decision-makers. As we will highlight below, a greater emphasis on such deviations
18
19
can enable scholars to further understand the performance implications of firms’ governance
20
21
22 choices.
23
24 There are also unexplored opportunities regarding opportunism to integrate insights into
25
26 TCT from behavioral economics and from different strands of psychology. As noted above, many
27
28
29 critics take issue with the somewhat pessimistic view of human nature implied in the opportunism
30
31 assumption in TCT. Yet, there is substantial evidence from work in behavioral economics and
32
33 psychology that people place different values on objects, and have systematically different
34
35
36
perspectives on events, based on their position relative to the object (e.g., an asset that one already
37
38 owns or is considering procuring) or event (e.g., what side of a disagreement one is on) that is
39
40 being considered. For example, prospect theory has highlighted the salience of loss-aversion,
41
42
which among other things results in an “endowment effect,” in which individuals tend to value
43
44
45 items that they already have more highly than is “rational” (Kahneman, Knetsch, & Thaler, 1990).
46
47 Alternatively, “self-serving” biases may systematically lead individuals to overvalue their
48
49 contribution, or the economic harm they face, and undervalue that of others – put differently,
50
51
52 systematically alter views of fairness (Roth & Murnighan, 1982). To illustrate, Loewenstein and
53
54 colleagues (Loewenstein, Issacharoff, Camerer, & Babcock, 1993) conducted an experiment in
55
56
57
58 12
59
60
Page 13 of 90 Academy of Management Annals

1
2
3 which they provided all participants with identical testimony from an actual court case, in which
4
5
6 an injured motorcyclist sued the insurer of the automobile driver who had collided with him. The
7
8 only difference was that half of the participants were told at the start of the experiment that they
9
10 were the plaintiff, while the other half were told that they were the defendant. Participants were
11
12
13
asked to submit their assessment of a fair settlement and their prediction of what the judge would
14
15 rule in the case. Their expectations diverged dramatically based on their putative identities, with
16
17 those assigned to be plaintiffs submitting “fair” and “expected” settlement figures that were nearly
18
19
double those submitted by those assigned to be defendants.
20
21
22 Why does this matter? The presence of such systematic biases may provide an alternative
23
24 route to predicting disagreements between economic actors as circumstances change, in a way that
25
26 does not rely on the conceptualization of opportunism as self-interest seeking with guile. In a world
27
28
29 with self-serving biases, or, for example, endowment effects that distort the subjective value of
30
31 already-owned assets, it is possible that what one transactor views as opportunistic behavior is
32
33 sincerely seen as fair behavior by the other. This brief discussion shows that integrating insights
34
35
36
from behavioral economics and psychology may yield alternative or additional ways of thinking
37
38 about the opportunism assumption that underlies the TCT logic, thus potentially mitigating some
39
40 of the concerns that have been raised about its original conceptualization.
41
42
Finally, opportunism has received substantial scrutiny in the literature on trust. Scholars
43
44
45 have proposed that trust can reduce transaction costs, primarily by reducing concerns about
46
47 opportunism (Anderson & Narus, 1990; Bromiley & Cummings, 1995). Trust can relate to a
48
49 party’s ability or to its benevolence/integrity (Ring & Van de Ven, 1992; Mayer, Davis, &
50
51
52 Schoorman, 1995), which reflects concerns about its bounded reliability or strategic opportunism,
53
54 respectively. Given empirical evidence showing that trust reduces transaction costs (Dyer & Chu,
55
56
57
58 13
59
60
 Academy of Management Annals Page 14 of 90

1
2
3 2003; Zaheer, McEvily, & Perrone, 1998), several scholars have proposed that trust – particularly
4
5
6 of the benevolence/goodwill type – can serve as a substitute to formal governance safeguards by
7
8 reducing or eliminating the threat of opportunism (e.g., Gulati, 1995; Gulati & Nickerson, 2008;
9
10 Lui & Ngo, 2004; an alternative view is that trust can complement effective contracting under
11
12
13
certain circumstances [Puranam & Vanneste, 2009]). As we further elaborate later, we see
14
15 substantial opportunity for continued engagement between the trust and TCT literatures. For
16
17 example, when TCT scholars have acknowledged differences in the level of trust, typically at the
18
19
societal level, this has been treated as a “shift parameter” that affects the optimal switching point
20
21
22 between governance modes (Williamson, 1991; Oxley, 1999). Advances in our understanding of
23
24 trust may allow scholars to unpack the “black box” of this shift parameter. At the same time, TCT
25
26 may be able to inform trust scholarship. Whereas much of the trust literature focuses on conditions
27
28
29 that enhance an individual’s propensity to trust others, a recent stream of work focuses explicitly
30
31 on “trust accuracy,” the proper placement of trust in those who warrant it and the withholding of
32
33 trust from those who do not (Schilke & Huang, 2018). As this literature seeks to flesh out the
34
35
36
situational factors that enhance trust accuracy (e.g., Schweitzer, Ho, & Zhang, 2018), TCT may
37
38 provide useful insights.
39
40 In Table 1, we provide an overview of our discussion in this section.
41
42
<<< INSERT TABLE 1 HERE >>>
43
44
45 THE KEY WILLIAMSONIAN CONSTRUCTS
46
47 Characteristics of transactions
48
49 Asset specificity. As we have noted, an asset is specific to a particular transaction if its
50
51
52 value in its next-best use and user (i.e., in a transaction with a different party) is less than its use
53
54 in the current transaction. The greater the difference between the value of an asset in its first-best
55
56
57
58 14
59
60
Page 15 of 90 Academy of Management Annals

1
2
3 and its next-best use, the more specific that asset is to the transaction. Williamson (1985) attached
4
5
6 particular significance to the role of asset specificity, arguing that it is the most important
7
8 determinant of governance. In the absence of asset specificity, economic actors can simply
9
10 terminate a transaction that has gone awry and find alternative partners with whom to transact.
11
12
13
Early empirical TCT research focused heavily on asset specificity, with numerous studies
14
15 finding systematic evidence that higher levels of asset specificity are associated with more
16
17 hierarchical governance (for an overview, see David & Han, 2004). The bulk of these studies
18
19
focused on the vertical integration decision, where asset specificity may be particularly important
20
21
22 due to how supply chain disruptions can harm a firm. As we describe below, other transaction
23
24 characteristics exhibit more prominent roles in studies of other governance decisions, such as
25
26 horizontal expansion.
27
28
29 Given the early emphasis on asset specificity, it is not surprising that this concept is the
30
31 most developed of the three key transaction characteristics. Williamson (1985) originally proposed
32
33 four distinct types of asset specificity: site specificity, physical asset specificity, human asset
34
35
36
specificity, and dedicated assets. In addition to these, several scholars have proposed other types
37
38 of asset specificity.6 For example, Masten, Meehan, and Snyder (1991) introduced the concept of
39
40 temporal specificity, which refers to investments in assets that require timely completion or
41
42
delivery to retain their value. Zaheer and Venkatraman (1994) discussed procedural asset
43
44
45 specificity, which refers to the degree to which workflows and processes are customized to the
46
47 transaction.
48
49
50 6 Site specificity refers to conditions that arise when an investment is made in close proximity to an exchange partner
51
and in which the setup and/or relocation costs are high. Physical asset specificity refers to assets that are mobile but
52
which have features that are designed for use within a specific transaction and that have lower values in alternative
53 uses. Human asset specificity relates to the degree to which the skills, knowledge, and experience of human assets
54 have value within a specific transaction. Finally, dedicated assets refer to the expansion of existing assets on behalf
55 of the exchange partner in ways that would leave the investor with excess capacity if the transaction were to be
56 terminated prematurely.
57
58 15
59
60
 Academy of Management Annals Page 16 of 90

1
2
3 Revisiting asset specificity. As Joskow (1988) points out, these different types of asset
4
5
6 specificity essentially capture the same set of underlying issues. Namely, they all center on settings
7
8 in which there are only a small number of actual or potential exchange partners, transactions
9
10 involve lock-in or dependence that is often asymmetric, and there are high incentives for
11
12
13
opportunistic behavior.7 Nevertheless, considering the variety of forms that asset specificity takes,
14
15 further conceptual refinement of each individual type can be useful. For example, whereas a
16
17 physical asset can be legally bound to an owner and – since it has no volition of its own – will
18
19
work equally well regardless of governance, a human “asset” has the freedom to leave an employer
20
21
22 (in most parts of the world) and has the discretion to work hard, shirk, or divert effort in various
23
24 directions. As we note below, many TCT studies present a generic argument pertaining to asset
25
26 specificity without considering the institutional details of their setting; we believe that future TCT
27
28
29 studies can benefit from introducing greater precision about how particular types of asset
30
31 specificity drive governance choices.
32
33 Furthermore, advances in the market design literature (e.g., Roth, 2008) may allow
34
35
36
additional conceptual refinement of the asset specificity construct and identify circumstances in
37
38 which hold-ups may arise even without dependence on conventional specific assets. A key feature
39
40 of a well-functioning market is “market thickness,” where a thick market maps closely onto TCT’s
41
42
notion of an exchange for which there are many potential partners. In this literature, thin markets
43
44
45 may arise for reasons other than asset specificity. For example, the digital transformation has had
46
47 a profound impact on competitive strategy and market structure. Of particular note is the idea that
48
49 network effects in two-sided markets (usually based around technological platforms) frequently
50
51
52 lead a market to “tip”; although many platforms may compete, it is often inevitable that only a
53
54
55
56 7 As we note below, there are other factors besides asset specificity that may generate a similar set of issues.
57
58 16
59
60
Page 17 of 90 Academy of Management Annals

1
2
3 small number survive (Arthur, 1994; Hagiu, 2014). Thus, a firm may initially have a large number
4
5
6 of platforms to work with, but then find its position fundamentally transformed into a small-
7
8 numbers situation, without itself having made investments in transaction-specific assets. As TCT
9
10 scholars work to extend the theory to platforms and ecosystems (as we recommend in our later
11
12
13
discussion in the section on the future of TCT), this is an interesting area for further research.
14
15 We conclude the discussion of asset specificity by highlighting three common empirical
16
17 challenges in the literature. First, many TCT studies that employ secondary-source data rely on
18
19
R&D or marketing intensity to proxy for the level of asset specificity. There are two concerns
20
21
22 arising from this. First, it is not clear that such intensity measures accurately reflect the level of
23
24 asset specificity for a firm. Although these studies implicitly or explicitly assume that high R&D
25
26 expenditure generates technological skills or assets that are specialized to a particular transaction,
27
28
29 and/or that high marketing expenditures reflect customer-facing skills or assets that are useful for
30
31 a narrow set of transactions, it is rarely clear why this is necessarily the case (Brouthers & Hennart,
32
33 2007).8 Second, even if we assume that these measures do indeed reflect specialized assets, many
34
35
36
studies use industry-average measures of R&D and marketing intensity, rather than firm-level
37
38 measures. It is true that data at the individual firm level may not always be available; however, the
39
40 use of aggregate industry measures reduces measurement precision at the very least and may also
41
42
lead to biased findings (e.g., Garrett, 2003). Hence, we strongly encourage scholars to collect data
43
44
45 at a lower level of analysis when this is feasible. More generally, given the vast number of studies
46
47 that have relied on firm-level accounting measures to proxy for asset specificity, further advances
48
49 in empirical research will benefit greatly from measuring asset specificity directly (examples of
50
51
52
53
54 8 In fact, the resource-based view has typically looked at these same measures as proxies of a firm’s ability to diversify
55 into new industries that benefit from modestly fungible technological or marketing resources (Montgomery &
56 Hariharan, 1991; Silverman, 1999).
57
58 17
59
60
 Academy of Management Annals Page 18 of 90

1
2
3 this are Stuckey [1983] and Hennart [1988a]), which is ideally done at the transaction level rather
4
5
6 than the firm- or industry-level.
7
8 Second, although TCT research has benefited tremendously from the many studies that use
9
10 primary/survey data to measure asset specificity, there is room to improve these measures as well.
11
12
13
For example, in several studies (e.g., Klein, Frazier, & Roth, 1990), survey questions assess the
14
15 amount of investment made to support a given transaction, but not the extent to which these
16
17 investments are specific, i.e., whether, and how much, the investment would have residual value
18
19
outside of the current transaction. Although there may well be a positive correlation between the
20
21
22 absolute level of investment and its asset specificity, scholars can assess this more precisely – and
23
24 could even measure the amount of quasi-rents at risk, as is sometimes done in labor economics
25
26 (Abowd & Lemieux, 1993; Guertzgen, 2009) – by collecting data on the value of the investment
27
28
29 outside its current use, i.e., its residual value. Such efforts would get closer to the essence of the
30
31 asset specificity construct. We thus encourage scholars to design surveys with the precise
32
33 theoretical construct more firmly in mind.
34
35
36
Third, the vast majority of TCT studies focus on the asset-specific investments by one
37
38 exchange partner, but do not consider the asset-specific investments made by the other partner.
39
40 This can be a critical omission, since the other partner presumably also must agree to the
41
42
governance mode selected for the transaction, and presumably has a preferred governance mode
43
44
45 based on its own level of asset-specific investments. Further, recent TCT research has highlighted
46
47 the role of “credible commitments,” or the reciprocal investment in transaction-specific assets by
48
49 both partners (so that both have something to lose if the transaction falls apart), as a mechanism
50
51
52 for supporting exchange (Williamson, 1983; Ahmadjian & Oxley, 2005). Given this, the focal
53
54 partner’s commitment to the transaction, in terms of making asset-specific investments, might pave
55
56
57
58 18
59
60
Page 19 of 90 Academy of Management Annals

1
2
3 the way for the other partner to make similar investments. Failing to incorporate the asset
4
5
6 specificity of the other partner – as a control variable, a hypothesized variable of interest, or
7
8 otherwise, in the research design – may generate an omitted variable bias. While some studies have
9
10 considered the asset specificity of both partners (e.g., Ganesan, 1994; Poppo, Zhou, & Li, 2016;
11
12
13
McEvily, Zaheer, & Kamal, 2017), future research will benefit from more systematically using
14
15 this comprehensive, integrated approach whenever data availability makes it feasible.
16
17 Uncertainty. Given Williamson’s emphasis on asset specificity, it is not surprising that
18
19
other transaction characteristics received less attention in early TCT research. In particular,
20
21
22 relatively little effort was made to conceptualize uncertainty or to distinguish among its different
23
24 forms. Williamson himself offered an oscillating view of uncertainty. In early work, Williamson
25
26 (1973: 318) focused on environmental uncertainty as a condition that would necessitate unforeseen
27
28
29 adaptation: “The effects of uncertainty on economic behavior are extensive and pervasive…. Of
30
31 particular interest to us here is that, inasmuch as a full set of contingent claim markets is infeasible
32
33 (by reason of bounded rationality), adaptive, sequential decision-making procedures need to be
34
35
36
devised. Vulnerable as market exchange is to opportunism in these circumstances, hierarchical
37
38 forms of organization are apt often to be favored.” By the next decade, however, Williamson
39
40 (1981) took pains to distinguish between non-strategic uncertainty and behavioral uncertainty.
41
42
Building on Koopmans’s (1957) distinction between uncertainty due to lack of knowledge about
43
44
45 potential states of nature (“primary” uncertainty) and uncertainty due to lack of knowledge about
46
47 the actions of other actors (“secondary” uncertainty) – which he interpreted as “non-strategic”
48
49 actions of others – Williamson proposed a third category, called behavioral uncertainty, which
50
51
52 relates to an inability to predict the opportunistic actions of others. For Williamson (1981),
53
54 behavioral uncertainty was the key characteristic that mattered. Yet by the next decade,
55
56
57
58 19
59
60
 Academy of Management Annals Page 20 of 90

1
2
3 Williamson (1991) had returned to his original conception of environmental uncertainty as the
4
5
6 relevant transaction characteristic.
7
8 Why this retrenchment? One interpretation might be that Williamson’s formulation of
9
10 behavioral uncertainty largely overlaps with the assumption of opportunism itself.9 If there is a
11
12
13
general distribution of propensity for opportunistic behavior among potential transaction partners,
14
15 then there is a latent level of behavioral uncertainty in all transactions. If so, then the relevant
16
17 variation across transactions is not behavioral uncertainty (which is always lurking) but
18
19
environmental uncertainty, since greater environmental uncertainty increases the likelihood that
20
21
22 an opportunistic actor will find a pretext to engage in unsavory behavior. Another interpretation
23
24 might be that, given its focus on vertical integration, traditional TCT accordingly privileged the
25
26 role of asset specificity (which may be particularly relevant for supply chain transactions), such
27
28
29 that uncertainty’s main role was to exacerbate the contracting problems associated with asset
30
31 specificity. As we discuss below, scholars who have extended TCT to horizontal expansion and
32
33 the multinational enterprise have re-invigorated and further dimensionalized the analysis of
34
35
36
behavioral uncertainty as a stand-alone concept.
37
38 Revisiting environmental uncertainty. Numerous scholars have followed the path of
39
40 assuming a background level of behavioral uncertainty, and consequently have focused on the
41
42
explication and measurement of environmental uncertainty. These studies typically follow
43
44
45 Williamson’s proposition that environmental uncertainty has a conditional effect on governance
46
47 rather than a direct one: “an increase in parametric uncertainty is a matter of little consequence for
48
49 transactions that are nonspecific…[but] whenever assets are specific in nontrivial degree,
50
51
52
53
54 9However, others have noted that even though behavioral uncertainty and opportunism are related, there can also be
55 sources of behavioral uncertainty that are less welded to opportunism. We will detail this view and the corresponding
56 sources of behavioral uncertainty in the next section.
57
58 20
59
60
Page 21 of 90 Academy of Management Annals

1
2
3 increasing the degree of uncertainty makes it more imperative that the parties devise a machinery
4
5
6 to ‘work things out’” (Williamson, 1985: 59-60). This view has been adopted and corroborated
7
8 empirically by several scholars (e.g., Leiblein, Reuer, & Dalsace, 2002; Lu & Hebert, 2005;
9
10 Cuypers & Martin, 2007). Leiblein (2003) concludes that empirical findings in the TCT literature
11
12
13
are generally consistent with this view.
14
15 At the same time, some scholars have postulated a direct effect of environmental
16
17 uncertainty on governance choice, and argued that such uncertainty makes it more difficult, if not
18
19
impossible, to contractually specify ex ante the circumstances surrounding an exchange. For
20
21
22 example, Walker and Weber (1984, 1987) looked at the impact of technological and demand
23
24 uncertainty on make-versus-buy decisions. However, the results of the studies that investigate a
25
26 direct effect of environmental uncertainty are mixed and inconclusive (Cuypers & Martin, 2007).
27
28
29 In addition, it is often unclear whether the direct effects of environmental uncertainty on
30
31 governance choice – when such effects are indeed found – are attributable to differences in
32
33 transaction costs, or rather can be explained by other theoretical approaches in which
34
35
36
environmental uncertainty plays a prominent role (e.g., real option theory). In sum, the literature
37
38 has clearly established that environmental uncertainty has the predicted joint effect with asset
39
40 specificity. In contrast, there is little systematic evidence that environmental uncertainty also has
41
42
a direct effect on transaction costs, and consequently about how it might affect governance choice.
43
44
45 This is not surprising considering that Williamson (1985) explicitly stated that environmental
46
47 uncertainty should have no direct effect on governance choices. Therefore, we suggest that
48
49 researchers who wish to continue this line of research, first and foremost, aim to extend and refine
50
51
52 the TCT logic in order to theoretically derive a direct effect of environmental uncertainty; or,
53
54 alternatively, seek to integrate TCT with complementary theories that predict such a direct effect.
55
56
57
58 21
59
60
 Academy of Management Annals Page 22 of 90

1
2
3 Revisiting behavioral uncertainty and appropriability. Some scholars have grappled
4
5
6 explicitly with the notion of behavioral uncertainty, often extending it beyond Williamson’s
7
8 somewhat narrow emphasis on strategic behavior alone. For scholars who place more emphasis on
9
10 bounded rationality and less on opportunism, Koopman’s (1957) “secondary uncertainty” – the
11
12
13
inability to predict others’ non-strategic behavior – may also affect forms of organization. For
14
15 example, the presence of irreducible knowledge differences between exchange partners (e.g.,
16
17 Conner & Prahalad, 1996) or the existence of bounded reliability (e.g., Verbeke & Greidanus,
18
19
2012) might be logically sufficient on their own to influence governance choice (although Foss
20
21
22 [1996a, 1996b] offers a contrary view).
23
24 Our review of the literature indicates that many scholars have conceptualized and
25
26 operationalized behavioral uncertainty in ways that encompass both Koopmans’s (1957)
27
28
29 secondary uncertainty concept and Williamson’s (1985) narrower focus on uncertainty of a
30
31 strategic kind. In addition, some scholars have conceptualized behavioral uncertainty in alternative
32
33 ways. Rindfleisch and Heide (1997) and Geyskens, Steenkamp, and Kumar (2006) both view
34
35
36
behavioral uncertainty as a performance evaluation problem, “that is, difficulty in ascertaining ex
37
38 post whether contractual compliance has taken place” (Geyskens et al., 2006: 521).
39
40 The international business literature on the multinational enterprise has been especially
41
42
innovative at highlighting and dimensionalizing behavioral uncertainty. This stems largely from
43
44
45 this literature’s focus on horizontal integration, rather than vertical integration (Dunning &
46
47 Lundan, 2008). Horizontal integration occurs when a firm integrates into production of a broader
48
49 range of goods and services (i.e., expansion into new product markets, or expansion into new
50
51
52 geographies with existing products). Whereas vertical integration largely turns on the dependence
53
54 across different stages of production, which highlights challenges of transaction-specific assets,
55
56
57
58 22
59
60
Page 23 of 90 Academy of Management Annals

1
2
3 horizontal integration is more commonly motivated by contracting difficulties caused by
4
5
6 information asymmetry, measuring costs, or other sources of behavioral uncertainty. For example,
7
8 information asymmetry plays a crucial role in the market for intangible assets (Hennart, 1982,
9
10 2015a). For markets to be efficient, buyers and sellers must possess the same information. In the
11
12
13
case of knowledge exchange, this is frequently not the case, since buyers often do not have precise
14
15 knowledge of what they are buying (Arrow, 1962). Similarly, although in principle a high-
16
17 reputation firm can license its brand to a firm in another country, it is sometimes difficult for the
18
19
firm to monitor the care with which the licensee treats the brand. As a result, the market for
20
21
22 intangible assets will sometimes break down, and firms will opt for a hierarchical solution rather
23
24 than a market based one (e.g., Davidson & McFetridge, 1982; Teece, 1985; Hennart & Park, 1994).
25
26 We will revisit this issue below when discussing work on appropriability. Our brief discussion in
27
28
29 this paragraph underscores the importance of factors other than asset specificity, such as
30
31 information asymmetry.
32
33 Further, the focus on multinational activity in international business research also
34
35
36
highlights cross-national differences that introduce variation in behavioral uncertainty. Thus, the
37
38 different phenomenological focus in international business research invites a more central role for
39
40 behavioral uncertainty, including the possibility of a direct effect on governance, rather than only
41
42
a joint effect with asset specificity. Consequently, this literature has explored a wide range of
43
44
45 sources of behavioral uncertainty. For example, numerous studies examine cultural distance across
46
47 countries, arguing that such distance increases behavioral uncertainty and thereby impacts
48
49 governance choices (for an overview see Tihanyi, Griffith, & Russell [2005] and Beugelsdijk et
50
51
52 al. [2018]). Some of these studies have proposed that cultural distance makes it harder ex ante to
53
54 understand and predict an exchange partner’s behavior (e.g., Anderson & Gatignon, 1986;
55
56
57
58 23
59
60
 Academy of Management Annals Page 24 of 90

1
2
3 Maseland, Dow, & Steel, 2018), while others have argued that cultural distance increases ex post
4
5
6 monitoring costs (e.g., Fladmoe-Lindquist & Jacque, 1995). However, as Maseland, Dow, and
7
8 Steel (2018) point out, most studies are not precise about how cultural distance creates behavioral
9
10 uncertainty.
11
12
13
In addition to cultural distance, several other factors have also been considered. Cuypers,
14
15 Ertug, and Hennart (2015) look at ownership levels in acquisitions, and argue that linguistic
16
17 distance between exchange partners and their level of English proficiency creates information
18
19
asymmetry and thereby behavioral uncertainty, and that this results in acquirers’ taking a larger
20
21
22 equity stake in targets. While Maseland and colleagues’ (2018) call for greater precision regarding
23
24 behavioral uncertainty was made with specific reference to cultural distance, it can also be applied
25
26 to the other factors that have been proposed. As examples of such precision, Henisz (2000) and
27
28
29 Henisz and Macher (2004) demonstrate that variation in countries’ political institutions affects the
30
31 level of “political hazards” faced by a multinational firm, and consequently influences the
32
33 decisions of whether and by what mode to invest in a country.
34
35
36
It is not only international business scholars who have provided valuable insights into the
37
38 factors that might drive behavioral uncertainty. As the knowledge-based economy has grown, a
39
40 cadre of (mostly strategy) scholars have devoted increased attention to innovation, technological
41
42
knowledge, and other intangible assets. This has led to the introduction of another important factor
43
44
45 that drives behavioral uncertainty, and thus governance: appropriability. Appropriability has been
46
47 discussed in the context of intellectual property and defined as the degree to which an economic
48
49 actor can protect its knowledge from leakage to other parties (Teece, 1986). This is a special case
50
51
52 of a more general requirement for market exchange, which is that market coordination – which
53
54 works through the exchange of ouputs – is only possible if these outputs are protected by property
55
56
57
58 24
59
60
Page 25 of 90 Academy of Management Annals

1
2
3 rights (Hennart, 1982). While some types of intangibles benefit from strong property rights and
4
5
6 hence can be traded on markets (Arora & Gambardella, 2010; Gans & Stern, 2010; Silverman
7
8 2019), this is not always the case, and the exchange of knowledge with poorly defined and
9
10 protected property rights will be afflicted by appropriability hazards. As foreshadowed earlier, a
11
12
13
key challenge stems from Arrow’s (1962) paradox of information: a seller of knowledge cannot
14
15 credibly convey its value unless she reveals the knowledge to the potential buyer, but the buyer
16
17 has no need to pay to acquire the knowledge once it is revealed.
18
19
The patent system is meant to solve this problem by making knowledge public, while
20
21
22 giving patentees property rights in the patented knowledge. Although such a system results in
23
24 strong appropriability for some technologies and in some countries, therefore making market
25
26 exchange of knowledge possible, the protection afforded by patents is imperfect – because patents
27
28
29 can only cover explicit knowledge, because patentees must shoulder the costs of enforcing their
30
31 own patents, and because the efficiency of public patent enforcement varies across countries
32
33 (Levin et al., 1987; Veugelers & Cassiman, 1999). Consequently, a strong “appropriability regime”
34
35
36
for intellectual property – e.g., a patent and trademark system that establishes and records property
37
38 rights and courts that provide their effective enforcement – can address this concern (Teece, 1986),
39
40 allowing markets for knowledge and reputation embedded in trademarks to function (Hennart,
41
42
1982; 2015a) Put differently, strong intellectual property protection can reduce the risk of
43
44
45 behavioral uncertainty in much the same way that reduced environmental uncertainty reduces the
46
47 risk of behavioral uncertainty.
48
49 Studies that incorporate appropriability have typically found that the stronger the
50
51
52 appropriability regime, the less likely hierarchy will be used to govern a transaction, as predicted
53
54 by TCT. For example, in a study of technology-driven diversification, Silverman (1999) finds that
55
56
57
58 25
59
60
 Academy of Management Annals Page 26 of 90

1
2
3 firms are more likely to exploit their technological assets by diversifying into a given industry
4
5
6 when that industry is characterized by weak patent protection, and infers that these firms rely on
7
8 licensing contracts to leverage these assets in industries characterized by strong protection. Oxley
9
10 (1999) finds that when U.S. firms engage in international R&D alliances, they employ more
11
12
13
hierarchical governance mechanisms when their partners come from countries with weaker patent
14
15 protection. This work highlights that resource characteristics, other than transaction-specificity,
16
17 might also be important drivers of behavioral uncertainty.
18
19
Based on these insights from the international business literature and the literature on
20
21
22 innovation and technological knowledge, we see potential for substantial benefits from a
23
24 systematic categorization of the factors that have been linked to behavioral uncertainty. If such a
25
26 categorization is done in a conceptually coherent manner, it is likely to encourage future research
27
28
29 to be more precise in its theorizing, improve our understanding of the behavioral uncertainty
30
31 concept itself, as well as its effects, and allow for better integration between TCT and other
32
33 theoretical perspectives. For example, the information economics literature (e.g., Spence, 1973,
34
35
36
2002; Arrow, 1984; Gibbons, Holden, & Powell, 2012) could inform us about the factors that
37
38 create information asymmetry ex ante or ex post, and thereby also influence behavioraluncertainty.
39
40 More sociological perspectives might help explain how dissimilarity or homophily (i.e., the
41
42
tendency to associate with similar others [Lawrence & Shah, 2020]) between exchange partners
43
44
45 can affect the perception of behavioral uncertainty (how predictable the other party’s behavior is,
46
47 e.g., Ertug, Gargiulo, Galunic, & Zou, 2018: 913), and institutional theory might explain how
48
49 contextual factors influence such uncertainty. Each of these ideas would have to be developed
50
51
52 substantially, but we believe that these brief examples illustrate that a categorization of theoretical
53
54 mechanisms is both feasible and valuable.
55
56
57
58 26
59
60
Page 27 of 90 Academy of Management Annals

1
2
3 Finally, another important issue in recent theoretical and empirical work is whether
4
5
6 scholars should emphasize behavioral uncertainty in an objective sense, or as it is perceived by
7
8 decision-makers. Perhaps because of TCT’s roots in economics, it is generally assumed (at least
9
10 implicitly) that decision-makers are able to accurately assess the level of behavioral uncertainty in
11
12
13
a transaction. Accordingly, behavioral uncertainty is usually conceptualized and measured as an
14
15 objective construct. However, in reality, managers make governance decisions based on their
16
17 perceptions of behavioral uncertainty. Accordingly, rather than assuming that these subjective
18
19
perceptions accord with objective reality, recent studies have increasingly discussed the relevance
20
21
22 of perceived behavioral uncertainty (e.g., Buckley & Chapman, 1998; Boersma, Buckley, &
23
24 Ghauri, 2003; Tsang, 2006). Such an approach is not inherently at odds with TCT, given that the
25
26 theory’s assumption of bounded rationality allows for decision-makers’ perceptions to deviate
27
28
29 from an objective assessment. Nevertheless, this does raise questions about which approach is
30
31 more suitable, or when one approach is more appropriate than the other.
32
33 We see room for both approaches. When the objective of a study is to establish how firms
34
35
36
should make decisions from a TCT perspective, and to verify if in fact decisions are made in the
37
38 predicted manner, we see objective measures as more appropriate. In contrast, when the objective
39
40 of a study is to explain deviations from TCT predictions, or to improve our understanding of how
41
42
managers make decisions that are based on TCT logic, perceptual measures might be more useful.
43
44
45 Specifically, although behavioral factors per se do not have to alter – or yield direct implications
46
47 about – how firms should make decisions, the work on heuristics, attention, risk preferences, and
48
49 biases (e.g., Kahneman et al., 1982; Ariely, 2008; Gigerenzer & Gaissmaier, 2011; Kahneman,
50
51
52 2011) could improve our understanding of how behavioral uncertainty is perceived and then
53
54 incorporated into decision-makers’ transaction cost analyses and the decisions they in fact make.
55
56
57
58 27
59
60
 Academy of Management Annals Page 28 of 90

1
2
3 This would better explain the kinds of decisions that are actually made, and why. Hence, both
4
5
6 approaches have their place in TCT research. We encourage scholars to devote more care and
7
8 effort to justifying and explaining their choice between perceived or objective behavioral
9
10 uncertainty, both conceptually and empirically. Although the distinction between objective and
11
12
13
subjective assessment is relevant for all transaction characteristics, we believe that it is most salient
14
15 for behavioral uncertainty, because this concept is more challenging to quantify and to objectively
16
17 assess, than asset specificity or frequency.
18
19
In sum, although most early TCT work emphasized asset specificity, more recently
20
21
22 behavioral uncertainty has been shown to be equally relevant, especially when TCT is applied
23
24 beyond its original vertical integration context. For research on the effects of behavioral
25
26 uncertainty to achieve its potential, it is important to further refine the concept and to be more
27
28
29 precise in terms of the mechanisms that link specific factors to behavioral uncertainty, and
30
31 consequently to firms’ governance choices.
32
33 Frequency. The third transaction characteristic is frequency, which refers to the extent to
34
35
36
which transactions recur. Williamson (1985) proposes that the overhead cost of hierarchical
37
38 governance should be easier to recover in the case of more frequently recurring transactions. As a
39
40 result, the likelihood of hierarchical governance should be higher for more frequently recurring
41
42
transactions than for less frequently recurring ones, all else equal. Several observers have noted
43
44
45 that, compared to asset specificity and uncertainty, frequency has received little attention in the
46
47 TCT literature (e.g., Rindfleisch & Heide, 1997). The extent of this dearth becomes clear when we
48
49 consider that two recent meta-analyses of TCT (David & Han, 2004; Geyskens et al., 2006) were
50
51
52 not even able to include frequency in their analysis, owing to the lack of studies that include this
53
54 construct. Furthermore, the few studies that have examined frequency have yielded mixed results.
55
56
57
58 28
59
60
Page 29 of 90 Academy of Management Annals

1
2
3 For example, Klein (1989) found a positive association between transaction frequency and the
4
5
6 degree of vertical control. However, Anderson and Schmittlein (1984) did not find an effect of
7
8 frequency on sales force integration (which represents a governance mode that is more hierarchical
9
10 than contracting with independent sales representatives).10 Given the paucity of studies that
11
12
13
investigate this construct, the mixed results in the small literature that does exist, and the more
14
15 skeletal theoretical rationale for this attribute, it may be time to redirect future efforts to better
16
17 understand the frequency-governance relationship.
18
19
Notably, it may be feasible to reframe an obstacle to the study of frequency – the challenge
20
21
22 of empirically distinguishing among Williamson’s (1985) proposed cost-amortization and other
23
24 mechanisms that relate to the recurrence of transactions – as a fruitful way to disentangle
25
26 alternative mechanisms. For example, repeated transactions are also associated with the
27
28
29 development of trust (e.g., Gulati, 1995), and with reductions in information asymmetry and the
30
31 development of routines (e.g., Reuer, Zollo, & Singh, 2002). Consequently, it is likely that these
32
33 mechanisms are confounded with the TCT mechanism in studies that rely on secondary data and
34
35
36
traditional regression approaches. One avenue forward might be to refine the empirical methods
37
38 to better identify the effects of frequency. For example, researchers might more directly capture
39
40 the cost implications of more-frequent transactions, which are typically measured as a latent
41
42
construct, or control more rigorously for the other mechanisms that are likely at play. Alternatively,
43
44
45 future research might consider exploring the effects of frequency in more controlled settings (e.g.,
46
47 Pilling, Crosby, & Jackson Jr., 1994) where frequency can be manipulated to allow for a more
48
49 precise examination of its effects, and to rule out alternative explanations.
50
51
52
53
54 10 Richman and Macher (2008: 6) note that studies that employ a continuous measure of frequency tend to find no
55 relationship between frequency and governance, while those that dichotomize frequency into one-time versus
56 recurring exchange tend to find the predicted relationship.
57
58 29
59
60
 Academy of Management Annals Page 30 of 90

1
2
3 To conclude, although transaction frequency is an inherent part of TCT, it has been largely
4
5
6 overlooked in both theoretical and empirical work. As a result, it is not known whether this factor
7
8 is determinative of transaction costs to the same degree as the other transaction characteristics in
9
10 the TCT framework, which have empirically been shown to be important drivers of transaction
11
12
13
cost and governance choices. Therefore, to the extent that future research grapples with frequency,
14
15 we encourage more systematic attempts to theoretically distinguish among cost-, trust-, and
16
17 routine-based effects, and to also explore novel empirical approaches to do so. It is possible that
18
19
such future research will find no relationship between frequency and governance; however, this,
20
21
22 too, is useful information, and we encourage researchers to report (and journals and editors to
23
24 publish) non-significant findings to avoid the “file-drawer” problem (Rosenthal, 1979), and if
25
26 applicable to implement proper methods to test for the existence of such null effects (e.g., Cuypers
27
28
29 & Martin, 2010; Quintana & Williams, 2018).
30
31 Governance modes and mechanisms
32
33 Markets, hierarchies, and hybrids. Williamson (1975) initially proposed two discrete
34
35
36
alternative governance modes that can be used to organize economic activity: markets and
37
38 hierarchies. These modes differ in terms of the mechanisms that enable them to govern exchange
39
40 and adaptation, with markets relying primarily on the price mechanism and hierarchies relying on
41
42
administrative control. In subsequent work, Williamson fleshed out intermediate modes within
43
44
45 these two polar forms, including long-term contracts (Williamson, 1979), relational contracting
46
47 (Williamson, 1979, 1981), and alliances or “hybrids” (Williamson, 1991). The consideration of
48
49 these other modes paralleled empirical TCT research that revealed a broad array of governance
50
51
52 mechanisms, including the lending of specialized assets to suppliers (Monteverde & Teece, 1982),
53
54 the matching of contract duration to the degree of asset specificity (Joskow, 1985), the signing of
55
56
57
58 30
59
60
Page 31 of 90 Academy of Management Annals

1
2
3 contracts that were demonstrably non-enforceable (Palay, 1985), reliance on strategic alliances
4
5
6 (Pisano, 1990; Oxley, 1997), and the exchange of credible commitments or “hostages” to facilitate
7
8 continuation of relationships (Heide & John, 1988; Anderson & Weitz, 1992; Ahmadjian & Oxley,
9
10 2005; Hennart, 1989). It also responded to criticism from non-TCT scholars that much economic
11
12
13
activity does not take place at either of the extreme poles of spot-market and within-firm (e.g.,
14
15 Powell, 1987; Stinchcombe, 1990). As a result, by the late 1990s, many TCT scholars recognized
16
17 a wide range of discrete governance modes along the spectrum from market to hierarchy, with
18
19
each mode being matched to a different set of transaction attributes (most commonly to different
20
21
22 levels of asset specificity or appropriability).
23
24 An alternative view of governance modes is provided by Hennart (1993), who notes that
25
26 the price mechanism rewards agents based on their outputs, while administrative control rewards
27
28
29 agents on their behavior, or inputs. The price mechanism thus elicits high effort but encourages
30
31 agents to cheat through quality-shading or other means, while administrative control elicits more
32
33 harmonious behavior but encourages agents to shirk.11 These methods thus incur different costs,
34
35
36
with the price mechanism requiring measurement of output to reduce quality-shading or other
37
38 forms of cheating, and administrative control requiring specification and enforcement of
39
40 behavioral constraints to reduce shirking.12 The relative cost of each organizing method will vary
41
42
with the specific characteristics of the transaction. With this background, Hennart (1993) then
43
44
45 proposes that the spectrum between market and hierarchy can include a large variety of governing
46
47 institutions that embody different combinations of price and administrative control. On the horns
48
49
50
51
52 11 Shirking is defined here as a failure to abide by the letter and spirit of one’s employment contract. Shirking can also
53 be defined as the difference in behavior between how employees behave and how they would behave if they were
54 running their own business (Hennart, 1993: 535).
55 12 Since it is impossible to perfectly monitor either behavior or output, there is also a residual cost of some irreducible
56 shirking under hierarchy and some irreducible cheating under the price system.
57
58 31
59
60
 Academy of Management Annals Page 32 of 90

1
2
3 of the dilemma between cheating and shirking, economic actors will judiciously choose an
4
5
6 appropriate combination of price- and administration-based mechanisms to govern their
7
8 transaction most efficiently. This approach suggests that a far greater set of transactions are
9
10 governed by institutions that are different from pure market and pure hierarchy than is traditionally
11
12
13
recognized. Such an approach also implies that there might be considerable variation within
14
15 modes, in terms of the bundles of governance mechanisms used, and that these bundles might
16
17 exhibit richer variation in governance modes than is recognized by the traditional market-hierarchy
18
19
continuum. For example, in a study of the design of contractual alliances, Reuer and Devarakonda
20
21
22 (2016) find that in some cases, i.e. when contractual alliances rely heavily on administrative
23
24 committees, contractual alliances might actually have more hierarchal features than equity joint
25
26 ventures, despite the fact that equity joint ventures are typically considered more “hierarchical”
27
28
29 than contractual alliances in the literature. This indicates the promise of more careful theorizing
30
31 and more in-depth empirical investigation of the properties of different governance alternatives (or
32
33 “institutions” in Hennart’s terminology). It is also worth highlighting that this view does not see
34
35
36
markets as the default option, to be replaced by hierarchy only when markets fail. Instead it
37
38 proposes that a full understanding of the choice of governance requires one to consider
39
40 simultaneously the factors that lead to market failure and those that lead to firm failure (Hennart,
41
42
1982, 1993).
43
44
45 Relational governance. In addition to the governance mechanisms we discussed above,
46
47 scholars have also proposed another form of governance, namely, relational governance. We will
48
49 briefly discuss the origins of the relational governance view and touch upon the alternative views
50
51
52 on how relational governance might fit within the TCT framework.
53
54
55
56
57
58 32
59
60
Page 33 of 90 Academy of Management Annals

1
2
3 Drawing on legal work by Macneil (1974, 1978), among other inspirations, Williamson
4
5
6 (1979) introduced the concept of relational contracting into TCT and considered it as a distinct
7
8 form of governance that relies on “private ordering” – resolution of disputes by the two parties to
9
10 an exchange, rather than by pursuing legal recourse via third-party courts. Relational contracting
11
12
13
emphasizes the ongoing nature of transactions and recognizes that these ongoing transactions are
14
15 embedded in relationships (for a sociological interpretation, see Granovetter, 1985). This view has
16
17 drawn the attention of numerous scholars, and a substantial literature on relational governance has
18
19
since emerged (e.g., Zaheer & Venkatraman, 1995; Dyer & Singh, 1998; Carson, Madhok, & Wu,
20
21
22 2006). This has also been a beneficial point of contact with non-TCT literatures. For example,
23
24 research on relational embeddedness has offered valuable insights, by demonstrating the ability of
25
26 relational embeddedness to enhance formal governance (e.g., Poppo & Zenger, 2002; Hoetker &
27
28
29 Mellewigt, 2009) and the ability of overall network structure to moderate the need for hierarchical
30
31 mechanisms in an alliance (Robinson & Stuart, 2007).
32
33 As the literature on relational governance expanded, some diverging views have emerged
34
35
36
that challenge TCT in ways that have substantial implications for work that seeks to combine the
37
38 logics of TCT and relational governance. Some scholars propose that network embeddedness or
39
40 trust can even obviate the need for TCT-style governance entirely and have proposed that networks
41
42
provide a way of organizing economic activities that is altogether distinct from markets and
43
44
45 hierarchies (e.g., Powell, 1990). Hence, they suggest that relational governance is a distinct
46
47 mechanism that does not lie on the continuum of governance structures proposed in TCT.
48
49 Alternatively, other scholars have taken a rather different, and less radical, view that allows
50
51
52 for easier theoretical integration of social mechanisms into TCT. They see the phenomena that
53
54 relational governance scholars focus on (for example, trust, embeddedness, or reputational
55
56
57
58 33
59
60
 Academy of Management Annals Page 34 of 90

1
2
3 concerns) as mechanisms that reduce behavioral uncertainty and thereby affect governance choices
4
5
6 (e.g., Argyres & Mayer, 2007; Malhotra & Lumineau, 2011). For example, Robinson and Stuart
7
8 (2007) study the network of alliances among U.S. biotechnology firms and find that alliances
9
10 between firms that are more proximate in the network, or that involve firms that are central in the
11
12
13
network, entail less hierarchical governance, ceteris paribus. They interpret this as evidence that
14
15 the ability to send and receive information throughout the network facilitates the use of a firm’s
16
17 cooperative reputation as a disciplining device, reducing the need for formal hierarchical control.
18
19
More generally, such variation in social mechanisms can act as a “shift parameter” that affects the
20
21
22 relative costs of governance modes (Williamson, 1991). The control-mechanism view is shared,
23
24 for example, by Hennart (2015b) who used guanxi in China as an example to demonstrate that
25
26 TCT can accommodate relation-based (or relational) governance, arguing that this kind of
27
28
29 understanding and discussion of relational and social mechanisms, i.e., within the TCT logic, also
30
31 makes it possible to evaluate the pros and cons of different types of social enforcement in an
32
33 objective manner.13 This view also highlights the relevance of TCT to the study of governance in
34
35
36
parts of the world where transactions are less commonly governed by contracts enforced by courts.
37
38 In Table 2, we provide an overview of our discussion in this section.
39
40 <<< INSERT TABLE 2 HERE >>>
41
42
THE LOGIC
43
44
45
46
47
48 13
49 Guanxi is the “the cultivating of personal relationships through the exchange of favors and gifts for the purpose of
obtaining goods and services, developing networks of mutual dependence, and creating a sense of obligation and
50
indebtedness” (Standiford & Marshall, 2000: 21). Relatedly, a firm can set up internal teams so that peer pressure can
51
be marshalled to curb employee opportunism, which is one well-known feature of Japanese management (Kenney &
52
Florida, 1993). Also, market actors may unilaterally internalize generalized morality – inculcated through schooling
53 and the family – and professional standards, taught through training and education (Ouchi, 1981; Hennart, 1991).
54 Differences in the extent of such internalization relate to the shift parameters that will alter the comparative costs of
55 different governance structures (Williamson, 1991), such that, for example, non-firm organization might be used for
56 higher levels of asset specificity in Japan than in the United States.
57
58 34
59
60
Page 35 of 90 Academy of Management Annals

1
2
3 Above, we discussed the essential building blocks of TCT. This discussion allows us now to
4
5
6 move on to examine the issues that pertain to the logic and how the logic can be extended.
7
8 A pluralistic view of TCT
9
10 Although TCT is typically most closely associated with Williamson (1975, 1985, 1991 and
11
12
13
1996), the sheer amount of research inspired by the theory has generated extensions and alternative
14
15 concepts that are not always recognized or used across different topic areas. Within the strategy
16
17 field, the most prominent extension has been the introduction of appropriability as a relevant
18
19
transaction characteristic, particularly for knowledge-based transactions (Teece, 1986; Oxley,
20
21
22 1997). Important work has also included a deeper recognition of the importance of credible
23
24 commitments, usually in the form of mutual investments in specific assets (Ahmadjian & Oxley,
25
26 2006 [although early work in this area came from marketing, e.g., Heide & John, 1988]), and a
27
28
29 rigorous assessment of interactions between formal and informal governance mechanisms (e.g.,
30
31 Poppo & Zenger, 2002).
32
33 In the international business field, key advances have emphasized the role of behavioral
34
35
36
uncertainty as a stand-alone transaction characteristic of consequence even in the absence of asset
37
38 specificity (Anderson & Gatignon, 1986; Hennart, 1982, 1993), as well as the range of
39
40 intermediate governance forms and the degree to which price and administrative mechanisms may
41
42
be judiciously combined to govern transactions (Gatignon & Anderson, 1988), and the question of
43
44
45 how a firm will choose to internalize a transaction – by building its own in-house operation or by
46
47 acquiring an existing operation (Anderson & Gatignon, 1986; Hennart & Park, 1993). The version
48
49 of TCT that is more commonly used in the international business literature has also been less
50
51
52 willing to assume that markets are the default governance structure, and has paid more attention
53
54 to outline the ways in which firms might “fail,” perhaps because the well-known challenges to
55
56
57
58 35
59
60
 Academy of Management Annals Page 36 of 90

1
2
3 multinational expansion (such as the liability of foreignness) highlight the challenges of expanding
4
5
6 firm boundaries. Presumably the TCT research in this field highlighted these issues because of the
7
8 classic international business-related phenomena under study: how to best exploit competitive-
9
10 advantage-generating assets in a new geographic market that has its own distinct institutional
11
12
13
background and culture. Given the wide range of mechanisms that multinational enterprises have
14
15 used to manage subsidiaries or overseas relations, especially when faced with high information
16
17 costs in monitoring those operations and with the prospect of negative spillovers to operations in
18
19
other countries if, say, a brand is damaged in one host country, it is not surprising that international
20
21
22 business scholars have devoted attention to these issues as they have worked to extend TCT
23
24 (Hennart, 1991). At the same time, strategy scholars could also benefit from more thoroughly
25
26 grappling with these ideas, particularly with respect to product-market diversification and to
27
28
29 reputational spillovers across product lines, which are relevant for strategy research.
30
31 The above extensions and alternative concepts have yielded valuable insights that have
32
33 substantially improved our understanding of the boundaries of the firm and firms’ governance
34
35
36
choices. However, several of these contributions have not widely diffused beyond their own fields
37
38 and streams of research. Considering such extensions allows us to think about opportunities
39
40 regarding how TCT, as a whole – with respect to its use in different areas of research – can be
41
42
advanced. Integrating and reconciling the logics underlying these extensions and alternative views
43
44
45 would lead to a more complete version of TCT. Figure 2 provides our sense of promising current
46
47 and future approaches to TCT.
48
49 <<< INSERT FIGURE 2 HERE >>>
50
51
52 A more dynamic TCT
53
54 TCT has been criticized for being static (e.g., Langlois, 1992; Zenger, Felin, & Bigelow,
55
56
57
58 36
59
60
Page 37 of 90 Academy of Management Annals

1
2
3 2011). Even though there has been progress in conceptually clarifying how TCT can be used in a
4
5
6 more dynamic way (e.g., Williamson, 1999), empirical research has rarely taken a dynamic
7
8 approach. This is an increasingly salient shortcoming of the TCT literature, since the contexts in
9
10 which many firms make decisions have become more dynamic and turbulent, and also given the
11
12
13
general trend toward more longitudinal, dynamics-sensitive research in the management field.
14
15 Accordingly, we discuss how TCT can be used in a more dynamic way and provide our perspective
16
17 of how future empirical research can address a number of questions that are particularly relevant in
18
19
today’s competitive landscape by taking a more dynamic approach.
20
21
22 We will start by elaborating on two dynamic, or intertemporal, issues within the existing
23
24 TCT framework. First, Williamson (1975) highlighted the importance of the “Fundamental
25
26 Transformation” in governance decisions. This transformation refers to the fact that the buildup of
27
28
29 asset-specific investments during contract implementation results in a shift from large-numbers to
30
31 small-numbers conditions and to the possibility of one party holding up the other at contract
32
33 renewal. An implication of this is that ex ante small-number conditions are not a necessary
34
35
36
condition when a governance choice is made, and the anticipation of this fundamental
37
38 transformation ex post should be driving governance choices ex ante. Hence, this is clearly an
39
40 inherently dynamic aspect in the TCT logic as outlined by Williamson. Related to this point, we
41
42
see three potential research opportunities. Incorporating the implications of this fundamental
43
44
45 transformation requires foresight from managers, as well as an accurate understanding of the
46
47 process, for them to make governance choices that are consistent with the TCT logic and
48
49 appropriate in the long run. Current research has not explicitly investigated the role played by
50
51
52 managers’ foresight and their understanding of the fundamental transformation in their governance
53
54 choices. In addition, this fundamental transformation is likely to be an endogenous process in some
55
56
57
58 37
59
60
 Academy of Management Annals Page 38 of 90

1
2
3 contexts. Hence, it would be useful to explore how firms’ ability to influence this process then
4
5
6 affects their own governance choices, as well as the governance choices of other firms. Finally, a
7
8 similar transformation might occur within hierarchies as well, which is a topic that has received only
9
10 limited empirical or theoretical attention (Argyres & Liebeskind, 1999). For example, employees’
11
12
13
incentives might be transformed over time in similarly systematic ways because of their
14
15 accumulation of firm-specific skills and knowledge. Thus, there are several research opportunities
16
17 by which scholars can refine and extend the dynamism inherent in Williamson’s “Fundamental
18
19
Transformation.”
20
21
22 Alternatively, several scholars have incorporated dynamics by studying the adaptation of
23
24 governance modes over time. Williamson (1991) argued that TCT can be the basis of a comparative
25
26 analysis that explains the adaptation of governance modes to changing circumstances. Following
27
28
29 this line of reasoning, Nickerson and Silverman (2003) found that U.S. trucking firms adapt their
30
31 governance choices in response to an exogenous shock to the conditions surrounding their
32
33 transactions, but at varying rates. Reuer and Arino (2002) examined whether or not changes in the
34
35
36
environment affect the decision to renegotiate the governance structure in alliances, finding
37
38 evidence that contractual terms are restructured to reduce misalignment caused by environmental
39
40 changes. Brahm and Tarzijan (2014) demonstrate that Chilean construction firms adapt their
41
42
governance structures and increased their extent of vertical integration after a legal shock that
43
44
45 increased contracting hazards.
46
47 This small body of work presents a number of opportunities for future research. First, most
48
49 studies have focused on discrete and substantial changes in governance mode (e.g., shifts from a
50
51
52 market-based or hybrid governance mode to a hierarchy-based governance). However, firms might
53
54 also make finer-grained adaptations to their governance modes to minimize transaction costs that
55
56
57
58 38
59
60
Page 39 of 90 Academy of Management Annals

1
2
3 arise from changes in the conditions surrounding the transaction. For example, in response to
4
5
6 contextual changes, firms might continue with a joint venture rather than move to a wholly-owned
7
8 subsidiary, but they might increase the levels of hierarchical control within the joint venture by
9
10 altering the ownership stakes and/or reshaping the board of directors (e.g., McQuade & Gomes-
11
12
13
Casseres, 1992). It would be worthwhile to investigate such finer-grained governance adaptations.
14
15 Second, most studies have focused on governance adaptations in response to external changes, such
16
17 as regulatory changes (e.g., Nickerson & Silverman, 2003). One reason for this is that an external
18
19
change simplifies research design: with an exogenous shock, one has a natural way to address
20
21
22 concerns about the endogeneity of governance choice. However, governance adaptations might also
23
24 occur due to internal changes. For example, the nature of investments might change over time and
25
26 become less specific, thereby triggering a need for governance adaptation (e.g., Hennart, 2009).
27
28
29 Accordingly, future research could also investigate how such internal factors lead to governance
30
31 adaptations. Finally, the adaptation process itself remains inadequately understood. Combining
32
33 insights from the literature on organizational adaptation (e.g., Ring & Van de Ven, 1994) with the
34
35
36
TCT literature might help us better understand when and how firms adapt their (misaligned)
37
38 governance modes. For example, Reuer, Zollo, and Singh (2002) show that prior experience can
39
40 facilitate governance adaptation in general. Accordingly, further studies could investigate how
41
42
experience in general, as well as different kinds of experience more specifically, could help the re-
43
44
45 alignment of governance modes that are misaligned from a TCT perspective. The identification of
46
47 other kinds of factors that facilitate or hamper realignment would also be relevant, both theoretically
48
49 and practically; as an example, Argyres, Mahoney, and Nickerson (2019) propose an approach that
50
51
52 integrates adjustment costs, transaction costs, and opportunity costs.
53
54 In sum, we believe that TCT is by no means inherently static and that it can accommodate
55
56
57
58 39
59
60
 Academy of Management Annals Page 40 of 90

1
2
3 more dynamic approaches. Even though only a modest number of studies have applied TCT in a
4
5
6 more dynamic way, our discussion and the examples above about future research indicate the
7
8 considerable payoffs in continuing these efforts and taking them further.
9
10 Performance implications
11
12
13
Management scholars have an inherent interest in understanding what drives firm
14
15 performance. While TCT explicitly discusses performance implications (e.g., Williamson, 1985,
16
17 1999), the literature has predominantly focused on explaining strategic choices, and relatively few
18
19
studies have looked directly at the performance implications of these choices (David & Han, 2004).
20
21
22 There is a clear opportunity for more TCT work that studies performance and that clarifies how the
23
24 theoretical logic pertains to performance.
25
26 TCT presumes that economic actors will enjoy performance benefits when they govern
27
28
29 transactions appropriately – or, put differently, firms whose transactions are not properly aligned
30
31 with appropriate governance structures will suffer performance consequences. The presumption that
32
33 firms with misaligned governance structures will suffer performance consequences relies “in a
34
35
36
general, background way on the efficacy of competition to perform a sort between more and less
37
38 efficient models and to shift resources in favor of the former” (Williamson, 1985: 22). As a
39
40 consequence of such competitive pressures, firms whose transactions are inappropriately governed
41
42
should perform worse, compared to those firms whose transactions are appropriately governed, such
43
44
45 that the former set of firms are pressured to adapt or risk being forced to exit a market.
46
47 The above implies that studying performance from a TCT perspective is both conceptually
48
49 and empirically more complex than is the case for many other theories. Namely, rather than thinking
50
51
52 in terms of how a given factor affects performance, studies that look at performance from a TCT
53
54 perspective need to first conceptually clarify and empirically identify misaligned and properly
55
56
57
58 40
59
60
Page 41 of 90 Academy of Management Annals

1
2
3 aligned governance structures, and only then compare these in terms of their performance
4
5
6 implications.14 However, such research comes with several empirical challenges (as we elaborate
7
8 below). We observe two broad approaches that researchers have used to empirically investigate the
9
10 performance implications of governance modes. One involves employing more sophisticated
11
12
13
empirical approaches, such as two-stage regression approaches (e.g., Anderson, Dekker, & Van den
14
15 Abbeele, 2017), often also taking advantage of external shocks (e.g., Hamilton & Nickerson, 2003).
16
17 The other is to focus on predicted tradeoffs in performance that are associated with a governance
18
19
choice, and then to seek evidence supporting or refuting these tradeoffs (without the presumption
20
21
22 that one mode is universally better than another). We consider these two approaches in turn.
23
24 Regarding the first broad approach, which employs more sophisticated empirical approaches,
25
26 Masten and colleagues (1991) use a two-stage-least-squares approach to measure transaction costs
27
28
29 for sea-vessel construction. Through interviews and surveys within one shipbuilder, they collect
30
31 information on the degree to which each of 74 components exhibits high asset specificity and
32
33 complexity, whether it is made in-house or outsourced, and, for the in-house components, the amount
34
35
36
of money spent governing the component’s production. Through the features of this econometric
37
38 approach, they are able to infer the transaction costs associated with the outsourced components, and
39
40 the expected costs had the outsourced parts been made in-house, or vice versa. Their results indicated
41
42
that transaction costs comprised 14 percent of total production costs, and misalignment would have
43
44
45 doubled these costs.
46
47 More commonly, scholars explore the effect of misalignment on profits or survival.
48
49
50
51 14 An example of a stream of research that has struggled with this issue is the literature on how multinationality affects
52
firm performance. Studies in this stream of work have typically looked at how levels of multinationality affect firm
53 performance (without considering whether a particular level of multinationality is in fact better or worse aligned with the
54 specific characteristics of the firm in question) (e.g., Contractor, Kundu, & Hsu, 2003; Lu & Beamish, 2004). Only a few
55 studies have conceptually emphasized the importance of considering whether the MNE’s level of multinationalilty is in
56 fact properly aligned or misaligned (e.g., Hennart, 2007, 2011) and empirically explored this (e.g., Powell, 2014).
57
58 41
59
60
 Academy of Management Annals Page 42 of 90

1
2
3 Nickerson and Silverman (2003) find that the over- or under-use of company drivers and company
4
5
6 trucks is associated with lower return on assets and lower likelihood of survival for motor carriers in
7
8 the post-deregulation U.S. trucking industry. Relatedly, Argyres and Bigelow (2007) find that
9
10 misalignment in production of engines is associated with lower survival for automakers – but only
11
12
13
after the “shakeout” stage of the industry life-cycle had begun. Gartenberg and Pierce (2017)
14
15 leverage the 2008 financial crisis to find that the vertical integration-performance relationship for
16
17 banks was moderated by the presence of strong corporate governance practices. Bruce and
18
19
colleagues (2019) take a different approach, relying on environmental stickiness rather than a shock.
20
21
22 In a study of the governance of private R&D contracts funded by the U.S. federal government, they
23
24 note that close project oversight is feasible for the government only when there are suitably expert
25
26 government personnel available. Using this factor as an instrument, they use two-stage methods to
27
28
29 measure the performance consequences of misaligned governance for projects, finding a substantial
30
31 penalty in terms of patents generated.
32
33 A number of scholars (e.g., Sampson, 2004; Powell, 2014; Anderson, Dekker, & Van den
34
35
36
Abbeele, 2017) have relied on another type of two-stage approach. They first determine whether a
37
38 governance choice is aligned or not with the context that surrounds the transaction, thereby arriving
39
40 at a measure of misalignment, and then regress their measure of misalignment against performance
41
42
in a second-stage. The measure of misalignment in these studies is typically the residual from the
43
44
45 first-stage regression (i.e. the difference between the observed governance choice and the estimated
46
47 – what would have been the properly aligned – choice). While this approach is more broadly
48
49 applicable, as it does not require a conveniently timed shock and sidesteps the often challenging task
50
51
52 of finding a reliable instrument (e.g., French & Popovici, 2011), it does assume that researchers are
53
54
55
56
57
58 42
59
60
Page 43 of 90 Academy of Management Annals

1
2
3 able to identify the appropriate governance choice.15 This highlights the importance of ensuring that
4
5
6 the first-stage model is properly specified (among other things, to limit the potential of omitted-
7
8 variable bias) and to carefully consider the validity of the misalignment measure.
9
10 Regarding the second broad approach, which focuses on tradeoffs associated with
11
12
13
governance choice, Novak and Stern (2008) note that a close reading of TCT implies that markets
14
15 should be particularly good at getting an arrangement right early on in an exchange, since parties
16
17 will spend time to specify important matters in a contract, whereas hierarchy should be particularly
18
19
good at managing change over time. They then collect longitudinal Consumer Reports data on 112
20
21
22 automobile models, as well as data on whether each of nine major components for each model was
23
24 outsourced or made in-house. Of particular note, once an automaker commits to internal (or external)
25
26 production for a model, it retains that mode for the life of the model. Consistent with the implication
27
28
29 of TCT, the authors find that models with more outsourced components outscore models with more
30
31 in-house components at the beginning of their lives, but that over time the score for heavily-
32
33 outsourced models deteriorates, whereas the score for insourced models does not, falling below the
34
35
36
insourced models after the third year. Thus, by looking at tradeoffs in a creative way, Novak and
37
38 Stern (2008) offer a valuable corroboration of the TCT-performance relationship without wading
39
40 into the endogeneity morass.
41
42
The above discussion highlights two important issues for researchers to consider when
43
44
45 deciding on a research design: endogeneity of the governance choice and survival bias. As Shaver
46
47 (1998) explains, it is reasonable to assume that a firm’s management team tries to make the best
48
49
50
51
52 15 As we also note below, a reasonable default assumption is that the firm’s decision-makers will make the best
53 alignment choice for their unique situation, so that what looks like misalignment to the researcher is plausibly due to
54 information that is unavailable to the researcher. At the same time, given the assumption of bounded rationality, some
55 managerial mistakes are to be expected. Recent empirical advances, such as propensity-score matching, that match
56 firms based on a wide range of observables, arguably go a long way to address this concern.
57
58 43
59
60
 Academy of Management Annals Page 44 of 90

1
2
3 choice given a firm’s idiosyncratic strength, weaknesses, opportunities, and constraints.16 As a result,
4
5
6 a researcher cannot easily examine the effect of a choice on performance, without having an adequate
7
8 strategy to deal with such endogeneity issues. Of the approaches we discussed above, the one that
9
10 relies on leveraging an exogenous shock might be particularly well-suited to deal with this issue. In
11
12
13
terms of the second issue, i.e. survival bias, TCT scholars have argued and shown that misaligned
14
15 firms or subsidiaries might be forced to exit a market (Argyres & Bigelow, 2007). This highlights
16
17 the importance of considering survival issues when sampling, or correcting for this bias empirically
18
19
with a selection stage. Despite these challenges, which are also relevant for other theories that relate
20
21
22 choices to performance, we are optimistic, given the growing set of econometric tools available to
23
24 deal with these issues.
25
26 In sum, there remains much exciting work to be done on the performance implications of
27
28
29 transaction alignment. For example, Forbes and Lederman (2010), Sampson (2004), and Anderson
30
31 and colleagues (2017) look at particular facets of performance, rather than overall profitability, and
32
33 find evidence consistent with TCT predictions in their investigation of operational performance in
34
35
36
the airline industry, innovative performance in R&D alliances, and residual risk in IT investments,
37
38 respectively. Hence, while the bulk of the work on performance in the TCT literature has focused on
39
40 profitability, these studies highlight the potential rewards of considering other types of performance
41
42
as well. Related to this, and inspired by Argyres and Bigelow’s (2007) work, future research could
43
44
45 investigate more thoroughly the conditions under which misalignment might lead to firm mortality,
46
47 versus those for which the performance consequences of misalignment are more forgiving; as de
48
49 Figueredo and Silverman (2012) show, relevant conditions may stem from governance decisions in
50
51
52
53 16Shaver (1998) compares the simple OLS approach to a two-stage approach that addresses endogeneity to demonstrate
54 that the “conventional wisdom” in the early 1990s – that foreign-market entry by greenfield plant performs better than
55 foreign-market entry by acquisition – was attributable to the fact that strong firms chose to enter by greenfield while
56 weak firms chose to enter by acquisition, and thus the choice-performance relationship was a spurious correlation.
57
58 44
59
60
Page 45 of 90 Academy of Management Annals

1
2
3 related industries. Sparked by Gartenberg and Pierce (2017), future research could also focus on
4
5
6 complementarities among governance features that jointly influence performance. Additionally,
7
8 continuing in the direction explored by Novak and Stern (2008), scholars could focus on subtle
9
10 differences in performance, not only to confirm or refute the predictions of TCT, but perhaps also to
11
12
13
test its explanatory power compared to that of other theories (for example, if property rights theory
14
15 does not suggest the same pattern of early versus late relative performance, then this would be a way
16
17 to distinguish between the implications of these theories). More generally, we see potential for more
18
19
work that improves our understanding of how firm and contextual factors might aggravate or
20
21
22 alleviate the negative performance implications of misalignment.
23
24 In Table 3, we provide an overview of our discussion in this section.
25
26 <<< INSERT TABLE 3 HERE >>>
27
28
29 THE FUTURE OF TCT
30
31 Above we provided an overview of the basic proposition of TCT, its underlying
32
33 assumptions, key theoretical constructs, and logic. We used this set of parameters to provide a
34
35
36
roadmap for navigating the vast TCT literature and to critically assess it. Throughout these
37
38 sections, we brought up a number of specific suggestions for future research. In this section we
39
40 identify three areas where TCT can be conceptually extended, and three areas where it can be
41
42
expanded to the study of important and recently emerging phenomena. We elaborate on these
43
44
45 extensions below, and provide an overview of this discussion in Table 4.
46
47 <<< INSERT TABLE 4 HERE >>>
48
49 Conceptual extensions
50
51
52 1. Further integration of strategy, international business, and institutional economics
53
54
55
56
57
58 45
59
60
 Academy of Management Annals Page 46 of 90

1
2
3 As described throughout the previous sections, strategy and international business
4
5
6 researchers have advanced TCT in ways that are relevant to their respective fields, but which can
7
8 also be productively combined. We believe that there are many more opportunities for such a
9
10 combination, which have the potential to lead to extensions in both fields. For example, consider
11
12
13
contracts that span national borders. A large body of TCT research explores domestic contracts,
14
15 demonstrating that contract terms are tailored to the key characteristics of transactions (e.g.,
16
17 Argyres, Bercovitz, & Mayer, 2007). For example, Reuer and Arino (2007) found that the level of
18
19
asset specificity is related to the complexity of contracts and the adoption of enforcement
20
21
22 contractual provisions. This research has devoted much less attention to how international factors
23
24 influence contract design, yet such work offers substantial opportunity to dimensionalize
25
26 contracting. Zhou and Poppo (2010) find that changes in the perceived legal enforceability of
27
28
29 contracts over time within China have an impact on the relationship between asset specificity
30
31 (among other factors) and contractual terms. Luo (2005) finds that institutional variation within
32
33 China influences the terms of contracts in international joint ventures. These studies highlight the
34
35
36
potential for future research to explore how heterogeneity in legal frameworks and in contract
37
38 enforceability across countries affect transaction costs and the design of contracts. In addition,
39
40 recent TCT work shows that linguistic differences between exchange partners influence
41
42
governance choices, for example how much ownership acquirers take in targets (e.g., Cuypers et
43
44
45 al., 2015). This underscores the potential value in investigating how language differences, both
46
47 between exchange partners and across countries, might influence contract design and effectiveness.
48
49 We note, however, that not all of the above-discussed views and extensions are
50
51
52 complementary, and in fact some of them even appear to be contradictory. For example,
53
54 Williamson’s (1991) and Hennart’s (1993) views on governance modes, and specifically on how
55
56
57
58 46
59
60
Page 47 of 90 Academy of Management Annals

1
2
3 to conceptualize hybrids, appear difficult to reconcile. Hence, another important area for future
4
5
6 research is to juxtapose such alternative views to see if such apparent contradictions can be
7
8 resolved, and if not, which conceptualization is preferable when. To be fruitful, these efforts need
9
10 to clearly lay out and then investigate rigorously the arguments and boundary conditions on respect
11
12
13
to each side. Such investigations would yield a better calibration (to the phenomena under
14
15 consideration) and more precise exposition (of the logic, constructs, and boundary conditions) of
16
17 each viewpoint, thereby making it more likely that the resolving of apparent contradictions and
18
19
proposed syntheses/juxtapositions in fact do yield both theoretical and empirical advances. Overall,
20
21
22 we strongly encourage more dialogue between different fields and streams of TCT research.
23
24 We also see potential to incorporate insights from other literatures in strategy, such as the
25
26 resource-based view and organizational learning. For example, Mayer and Argyres (2004)
27
28
29 emphasized the importance of experience in making optimal governance choices, and Argyres and
30
31 colleagues (2012) referred to firms’ having a governance capability. This small but growing
32
33 literature suggests that capabilities and firms’ experience play a positive role in terms of firms’ opting
34
35
36
for governance structures that are better aligned with the transactional characteristics. We see
37
38 opportunities in continuing this line of work by bringing in additional nuances from these literatures.
39
40 The work that links experience and capabilities with TCT has largely taken a static approach so far.
41
42
Therefore, introducing the notion of dynamic capabilities might facilitate a more dynamic
43
44
45 perspective, as such capabilities might help firms to identify the need to adjust misaligned
46
47 governance structures and to mobilize the necessary resources to do so. Moreover, most work has
48
49 implicitly assumed that the effects of experience and capabilities on governance choices are
50
51
52 uniformly positive. However, in the organizational learning literature it is well established that
53
54 experience can also have negative effects (e.g., Haleblian & Finkelstein, 1999). Hence, we see
55
56
57
58 47
59
60
 Academy of Management Annals Page 48 of 90

1
2
3 opportunities for future research to explore the possibility of such negative effects in the context
4
5
6 of governance choices as well.
7
8 Moving slightly farther afield, TCT has long drawn on economics, but can benefit from
9
10 further integration with two related theoretical lenses, namely information economics and
11
12
13
signaling theory. The appropriability literature in strategy and work on behavioral uncertainty in
14
15 international business have clearly highlighted the relevance of information asymmetry in TCT.
16
17 Incorporating additional insights from information economics will likely improve our
18
19
understanding of the different types and sources of information asymmetry that might drive
20
21
22 behavioral uncertainty. In addition, signaling theory (for a review, see Connelly et al., 2011) can
23
24 offer valuable insights on how behavioral uncertainty can be reduced. Specifically, it might
25
26 improve our understanding of the mechanisms that firms can use to reduce information asymmetry
27
28
29 and thereby also the level of (perceived) behavioral uncertainty their transaction partners face.
30
31 Hence, these two theoretical lenses have the potential to lead to a better conceptualization and
32
33 understanding of the behavioral uncertainty construct.
34
35
36
2. Link to sociology – trust/opportunism; formal/informal organization
37
38 A vibrant literature on trust explores the ways in which interorganizational trust develops
39
40 and the ways in which it can influence organizational arrangements. Definitions of trust commonly
41
42
include a willingness to “accept vulnerability based upon positive expectations of the intention or
43
44
45 behaviors of another” (Rousseau et al., 1998: 395).17 Studies of trust in interorganizational
46
47 arrangements find that it is positively associated with collaborative performance (Zaheer et al.,
48
49
50
51
52
53 17 Trust may be disaggregated into competence-based (I expect that you are able to keep your promise) and goodwill-
54 based (I expect that you are willing to keep your promise) (Lui & Ngo, 2004). It may also be disaggregated into
55 calculation-based (I make a probability-based risk assessment) and relation-based (I make a leap of faith) (Schilke &
56 Cook, 2015; Poppo et al., 2016; see McEvily [2011] for a hybrid construction of calculative and relational trust).
57
58 48
59
60
Page 49 of 90 Academy of Management Annals

1
2
3 1998; Dyer & Chu, 2000, McEvily, Perrone, & Zaheer, 2003),18 and can serve to alter the degree
4
5
6 of formal governance used in the relationship (Gulati, 1995; Zaheer & Venkatraman, 1995). Gulati
7
8 and Nickerson (2008) conceive of trust as a shift parameter that allows actors to reduce their
9
10 concern about opportunism, and consequently use a less formal governance structure than would
11
12
13
otherwise be necessary. Puranam and Vanneste (2009) propose that different aspects of trust may
14
15 either substitute for or complement formal governance, but the expectation either way is that
16
17 greater trust smooths relations and enhances performance.
18
19
A number of studies in this area implicitly assume that trust is well placed (Gargiulo &
20
21
22 Ertug, 2006).19 Consequently, a common assumption in the literature is that trust increases with
23
24 repeated interactions (e.g., Gulati, Lavie, & Singh, 2009). Theoretical models (e.g., Puranam &
25
26 Vanneste, 2009) tend to focus on the consistent building of trust, rather than its dissipation, perhaps
27
28
29 through a series of modest interactions that build confidence, among the partners, allowing them
30
31 to shift from calculative trust to a more faith-based, heuristic trust (McEvily, 2011). In such a
32
33 world, it is not surprising that trust would appear to be a strong antidote to opportunism.
34
35
36
Yet a small number of studies that explore trust in conflictual settings find evidence that
37
38 the trust-opportunism relationship is more nuanced than conventional wisdom might suggest.
39
40 Graebner (2009) examines the role of trust in acquisitions through eight case studies in which
41
42
technology-based entrepreneurial firms were acquired by larger suitors. She presents noteworthy
43
44
45 results: almost all of the executives from the acquiring firms engaged in deceptive practices,
46
47 ranging from making empty promises of bridge financing to dissuade targets from pursuing
48
49
50 18 Many studies rely on surveys that measure trust and performance at the same point in time. One concern about this
51
method is that, as long as an arrangement seems to be going well, parties are more likely to feel trust for each other;
52
this suggests a positive correlation between trust and performance in which trust does not necessarily cause good
53 performance.
54 19 E.g., Trust “is a powerful alternative to formal governance mechanisms that attempt to align incentives and control
55 opportunism through monitoring and sanctions….The greater the level of ex post trust in the exchange relationship,
56 the greater the benefits to the relationship will be” (Puranam & Vanneste, 2009: 11).
57
58 49
59
60
 Academy of Management Annals Page 50 of 90

1
2
3 venture financing or seeking alternative suitors (which would have raised the ultimate price of
4
5
6 acquisition), through misrepresentating their outside options and reservation prices during the
7
8 haggling over acquisition price, to making false commitments regarding post-acquisition behavior
9
10 (e.g., promising not to fire or relocate employees when they fully intended to do so post-
11
12
13
acquisition). Of particular note is the finding that these executives engaged in more deception
14
15 when they believed that the target executives were more trusting. Finally, some of the target
16
17 executives never realized that they had been deceived, even after the acquisition was consummated.
18
19
This study provides striking evidence of opportunism in action, and of the inability of
20
21
22 accomplished executives to always distinguish good faith from opportunistic behavior. It thus
23
24 suggests that trust may not always overcome opportunism, and reminds us that trust may open the
25
26 door for more severe opportunism (as noted by Granovetter, 1985: 491-492). This is buttressed by
27
28
29 laboratory-experiment evidence that high trust may damage performance by constraining people’s
30
31 willingness to monitor behavior (Langfred, 2004), and it may contribute to explaining why more
32
33 experienced parties capture more value in acquisitions (Cuypers, Cuypers, & Martin, 2017). An
34
35
36
intriguing line of trust research has begun to explore “trust accuracy” (Fetchenhauer & Dunning,
37
38 2010; Schweitzer et al., 2018; Schilke & Huang, 2018), and specifically the situational features –
39
40 such as interpersonal communication and the availability of monitoring mechanisms – that lead
41
42
parties to more accurately assess each other’s trustworthiness.
43
44
45 Relatedly, a few scholars have explored whether trust can repair relationships when they
46
47 break down. Jap and Anderson (2003) study buyer-supplier relationships and find that, after one
48
49 side suspects the other of opportunism, trust is a less useful lever through which to rectify the
50
51
52 relationship than is goal congruence or the presence of bilateral investments. Lumineau and
53
54 Malhotra (2011) find that the reservoir of trust between two parties can enable them to resolve a
55
56
57
58 50
59
60
Page 51 of 90 Academy of Management Annals

1
2
3 dispute rather than severing the relationship, and that this reservoir of trust is itself a function of
4
5
6 the contractual terms governing their collaboration.
7
8 For us, these studies open up a host of fascinating questions at the boundary of TCT and
9
10 trust. To what degree can parties reliably count on their perceptions of each other’s trustworthiness
11
12
13
to govern their relationship? How frequently is opportunistic behavior actually detected by a
14
15 trusting party? Does trust of another elicit reciprocal trusting (and trustworthy) behavior, or does
16
17 it create more temptation for opportunistic behavior? How do contracts affect these dynamics?
18
19
Lumineau and Malhotra (2011) find that control-related contractual provisions tend to reduce
20
21
22 goodwill trust, while coordination-related contractual provisions do not, and that this reduction in
23
24 goodwill trust increases the likelihood of ending a relationship after a dispute. To what extent can
25
26 we further understand the role of specific contractual terms in jointly affecting trust and
27
28
29 performance? Do coordination-related contractual provisions support the development of the goal
30
31 congruence that Jap and Anderson (2003) highlight? Can TCT inform the use of “situational
32
33 factors” that facilitate more accurate assessment of others’ trustworthiness? Finally, when conflict
34
35
36
arises, what affects a party’s perception of its partner’s benevolence and competence? In a
37
38 laboratory experiment, Harmon, Kim, and Mayer (2015) find evidence that subjects are more
39
40 forgiving of perceived incompetence than of perceived deliberate shirking. It is worth nothing that
41
42
their experiment exposed subjects to identical protestations of “I didn’t realize” from their partners,
43
44
45 while a specific contractual provision was varied so that the violation would be perceived to be
46
47 either of the letter or the spirit of the contract. Subjects perceived violations of the letter to be more
48
49 deliberate than violations of the spirit. Future work that further unpacks the link between formal
50
51
52 stipulations and trust in the case of conflict in collaborations would enhance our understanding of
53
54
55
56
57
58 51
59
60
 Academy of Management Annals Page 52 of 90

1
2
3 trust and opportunism, and, like the study by Harmon, Kim, and Mayer (2015), may also speak to
4
5
6 issues of framing (discussed in greater detail in the next subsection).
7
8 More generally, TCT has tended to focus on formal organization mechanisms (e.g.,
9
10 Argyres & Silverman, 2004). At the same time, a literature exists on the role of informal
11
12
13
organization (Gulati, 1995; Dyer & Singh, 1998). The last two decades have witnessed much work
14
15 on whether these are complements or substitutes (e.g., Poppo & Zenger, 2002; Mayer & Argyres,
16
17 2004). While much of this research has looked at these at a point in time, recent work has begun
18
19
to explore the relation between formal and informal organization (McEvily, Soda, & Tortoriello,
20
21
22 2014), and their influence on firm outcomes over time. For example, Argyres, Rios, and Silverman
23
24 (2020) study sharp shifts in the degree of centralization in firms’ R&D functions, and examine the
25
26 rate at which both the firms’ internal co-invention networks and their innovative outcomes change
27
28
29 afterwards. They find evidence that the shift in formal organization is followed by a gradual change
30
31 in co-invention patterns, and that the change in innovative outcomes is concurrent with the change
32
33 in co-invention. Such research helps to disentangle formal and informal mechanisms that affect
34
35
36
performance; beyond that, it helps to shed light on the factors that affect informal organization
37
38 itself. There are opportunities to further exploit changes in formal structure to understand their
39
40 effect on informal structure, and possibly vice versa.
41
42
3. Link to psychology/behavioral economics – context and framing
43
44
45 As our review indicates, the assumptions that underlie TCT clearly allow for the
46
47 incorporation of more behavioral perspectives. The recent incorporation of such behavioral
48
49 insights has benefited many theories (e.g., Felin, Foss, & Ployhard, 2015). However, despite being
50
51
52 well-positioned to leverage the shift toward a more behavioral perspective in many fields and
53
54 disciplines, TCT has not advanced sufficiently in this area to achieve its potential. Our review
55
56
57
58 52
59
60
Page 53 of 90 Academy of Management Annals

1
2
3 highlights a number of opportunities to integrate more behavioral perspectives in each of the
4
5
6 theory’s components. Insights from decision-making theory and social psychology (e.g.,
7
8 Kahneman et al., 1982; Ariely, 2008; Gigerenzer & Gaissmaier, 2011; Kahneman, 2011) could
9
10 inform us how heuristics, attention, risk preferences, and biases affect the extent to which decision-
11
12
13
makers evaluate the potential for opportunistic behavior. Considering these insights would also
14
15 improve our understanding of how behavioral uncertainty is perceived and incorporated into
16
17 decision-makers’ governance choices. We believe it is also crucial to incorporate behavioral
18
19
factors to make substantial advances in our understanding of why some firms deviate from the
20
21
22 governance choices they should make, as based on what TCT prescribes, and when these
23
24 deviations lead to lower performance. Likewise, work in this direction could also inform us why
25
26 some firms adjust their misaligned governance structures (or do this sooner) whereas others do
27
28
29 not. Overall, these examples highlight that future research could substantially benefit from
30
31 combining more behavioral perspectives with the TCT logic to explain actual governance choices
32
33 and their performance implications.
34
35
36
Extensions to new phenomena
37
38 1. Technological advance – platforms and governance
39
40 Recent years have witnessed the application of TCT insights to understand new phenomena
41
42
that leverage advances in communications technology including E-business (e.g., Amit & Zott, 2001;
43
44
45 Hennart, 2019), alliance constellations (e.g., Gomes-Casseres, 2003), innovation ecosystems (e.g.,
46
47 Adner & Kapoor, 2010), and technology platforms (e.g., Lehdonvirta, Kässi, Hjorth, Barnard, &
48
49 Graham, 2019). This modest body of work suggests that TCT is relevant and can indeed offer value
50
51
52 in explaining some of the important aspects of these phenomena. In addition to using TCT to
53
54 understand these new phenomena, conversely, one can also use these phenomena to extend and
55
56
57
58 53
59
60
 Academy of Management Annals Page 54 of 90

1
2
3 refine TCT. Notably, new business models such as platforms and ecosystems raise questions
4
5
6 pertaining to governance in a multi-party context, which encourage an extension of the TCT logic
7
8 beyond the usual dyadic level of analysis. Hence, there might be substantial theoretical rewards to
9
10 exploring new contexts and phenomena.
11
12
13
Hennart (2019) argues that a good starting point is understanding that most forms of e-
14
15 commerce have analogues in the pre-IT era. Hence, C2C platforms, such as e-Bay and e-harmony,
16
17 are brokers, putting in contact individuals in search of used household furniture or relationships,
18
19
analogous to classified advertisements, flea markets, and marriage bureaus. C2B job platforms, like
20
21
22 Indeed.com, are similar to government or university placement centers. Alibaba and other B2B sites
23
24 are electronic analogues of traditional brokers. B2C sites, like Booking.com and Expedia, provide
25
26 the same services as travel agents. Uber, Lyft and Ola offer taxi rides, just like traditional taxi
27
28
29 companies, and customers can choose between ordering from Amazon or buying from brick and
30
31 mortar retailers.
32
33 Similarly, platform-based ecosystems have analogues in the pre-Internet world, in situations
34
35
36
where scale and switching costs matter. Platforms such as Amazon or Apple emerge from differences
37
38 in optimal scale between the upstream and the downstream stages of a business. The optimal scale
39
40 of designing, manufacturing, and retailing a mobile phone is larger than that of designing apps. At
41
42
the same time, the attributes of apps are such that contracting between app developers and phone
43
44
45 makers is feasible; hence Apple can contract with app developers rather than vertically integrate into
46
47 app production. This is similar to the ethical drug business, where the optimal size of drug discovery
48
49 is smaller than that of drug marketing, with the result that the industry is characterized by a larger
50
51
52 number of drug discovery firms that contract with a smaller number of drug distributors (Big
53
54 Pharma). The power that mobile phone companies have over app developers (and Big Pharma has
55
56
57
58 54
59
60
Page 55 of 90 Academy of Management Annals

1
2
3 over drug discovery firms) derives from the fact that, because of scale economies, there are fewer
4
5
6 platforms than there are app developers. The same logic applies to large retailing platforms like
7
8 Amazon. Because of first mover advantages and customer switching costs, these firms have
9
10 obtained market power vis-à-vis the firms using their platforms. This is very similar to the case of
11
12
13
Sears when it was the dominant mail order distributor.
14
15 So, what is different? At least three items. First, network effects often magnify the scale
16
17 economies and first-mover advantages associated with a platform business (Arthur, 1994). This has
18
19
implications for the strategies that young platform businesses pursue to gain scale (Boudreau, 2017).
20
21
22 Of particular note for TCT, to the extent that these network effects create “winner-take-all” markets
23
24 that are ultimately dominated by a handful of entrenched incumbents, the difficulty of unseating an
25
26 incumbent platform owner can also magnify hold-up risks later in the platform’s life. TCT is likely
27
28
29 to be particularly useful at understanding how a new platform can try to attract partners, and how
30
31 those partners should anticipate future problems when contracting with the young platform. Second,
32
33 technology platforms typically have more information on both the profitability of the business
34
35
36
generated by the firms using their platforms and the behavior of its partners. For example, Amazon
37
38 appears to have used profit-related information to expropriate the profits of the partners using its
39
40 platform by replacing their products with its own (Zhu & Liu, 2018). And, although there are no
41
42
fundamental differences between the business model used by Uber and that used by franchisors
43
44
45 such as McDonalds,20 Uber has more power over its “franchisees” than does McDonalds. Whereas
46
47 McDonalds must send people to intermittently inspect its franchisees, Uber can do this remotely
48
49
50
51
52 20 In both cases, the minimum efficient scale at the upstream stage (format and product design and advertising for
53 McDonalds, platform for Uber) is greater than that at the downstream stage (restaurants for McDonalds, taxi service
54 for Uber). Subcontracting to independent agents (franchisees for McDonalds, drivers for Uber) makes sense in this
55 case, because the gains in effort and flexibility achieved by using independent contractors are greater than the potential
56 loss in quality shading that will inevitably result.
57
58 55
59
60
 Academy of Management Annals Page 56 of 90

1
2
3 and consistently through in-car cameras. Uber can also control driver behavior through the
4
5
6 navigation system drivers must use and keep them from cheating passengers through its payment
7
8 system.21 TCT should be particularly useful at analyzing the implications for the governance of
9
10 platform-based exchange of lower-cost monitoring of both behavior and output, Third, technology
11
12
13
platforms have an unusually high ability to bundle and to exclude (Boudreau, 2010; Hagiu &
14
15 Wright, 2015). Thus, it is much easier for Uber to cut off franchised drivers than it is for
16
17 McDonalds to terminate franchised outlets, or for Amazon to add its own, or third-party,
18
19
certification information than for Walmart (in their brick and mortar stores) to do so. From a TCT
20
21
22 perspective, this is a double-edged sword. On the one hand, platform owners can use these
23
24 advantages to elicit desired behaviors from partners (Rietveld, Seamans, & Meggiorin, 2020),
25
26 including enhanced coordination with other partners. At the limit, the platform owner may even
27
28
29 provide stronger institutions than the national government (Liu & Weingast, 2017). On the other
30
31 hand, potential partners should anticipate future dependence on the platform owner, and thus
32
33 demand strong contractual protections before joining. There are many opportunities to apply and
34
35
36
extend TCT analysis to these situations; our sense is that this extension will explain much of e-
37
38 business dynamics without the need to develop an entirely new theory of platform governance.
39
40 2. Technological advance – artificial intelligence, machine learning, and transaction costs
41
42
Here we illustrate how TCT can be applied to new phenomena by focusing on recent
43
44
45 technological trends, such as advancements in machine learning, artificial intelligence, big data,
46
47 and blockchain technology. We believe that each of these trends has the potential to affect the level
48
49 of behavioral uncertainty firms face, and the relative costs of different governance mechanisms.
50
51
52 For example, law scholars, legal practitioners, and management scholars who focus on contracting
53
54
55 21Note however that the introduction of new technology has not totally eliminated quality shading by drivers, which,
56 in the case of Uber, manifests itself in dangerous driving and assaults on customers.
57
58 56
59
60
Page 57 of 90 Academy of Management Annals

1
2
3 issues are starting to recognize that machine learning and artificial intelligence, combined with the
4
5
6 availability of big data on contracts, might pave the way for the partial, or even full, automation
7
8 of contract drafting, as well as of contract review and analysis (e.g., Mills, 2016; Betts & Jaep,
9
10 2017; Rich, Weber, & Bauman, 2020). Early applications of these ideas suggest that they could
11
12
13
lead to substantial efficiency gains and cost reduction in enforcement of contracting, thereby
14
15 making market-based transactions less costly. At the same time, machine learning and artificial
16
17 intelligence might lead to significant enhancements and expansion in the automation of tasks and
18
19
decision-making within firms (e.g., Jarrahi, 2018; Schneider & Leyer, 2019). Human resources
20
21
22 scholars and practitioners have acknowledged that these technologies have important implications
23
24 for monitoring and evaluating employees (e.g., Tambe, Cappelli, & Yakubovich, 2019; Kellogg,
25
26 Valentine, & Christin, 2019). Thus, advances in machine learning and artificial intelligence might
27
28
29 reduce shirking and the need for and cost of monitoring employees, as well as the potential for
30
31 opportunistic behavior more generally. This implies that the same technological trends that have
32
33 the potential to make market transactions more efficient also have the potential to have a similar
34
35
36
effect on hierarchies. Therefore, it is not only important that future research investigates the cost
37
38 implications of these technological trends for one particular governance mechanism or mode, but
39
40 also that such research considers more broadly how such technologies affect the relative cost of
41
42
different governance mechanisms (since it is these relative, comparative, assessments that allow
43
44
45 us to form predictions that are properly grounded in the TCT logic).
46
47 Another technological advance that might be relevant for TCT is blockchain technology.
48
49 While successful blockchain applications might be scarce at this moment, scholars have already
50
51
52 started to explore their potential (e.g., Schmidt & Wagner, 2019; Davidson, De Filippi, & Potts,
53
54 2018). Blockchains allow for immutable decentralized public ledgers and thus enable the keeping
55
56
57
58 57
59
60
 Academy of Management Annals Page 58 of 90

1
2
3 and sharing of records of past behavior in a distributed and decentralized way. Catalini and Gans
4
5
6 (2016) argued that this technology has the potential to lower transaction costs and improve the
7
8 efficiency of markets through costless verification and by reducing the need for costly
9
10 intermediation. Furthermore, as blockchain technology might make past opportunistic behavior
11
12
13
available as searchable public information, it also has the potential to reduce behavioral uncertainty
14
15 and to deter or reduce future opportunistic behavior.
16
17 3. Non-pecuniary phenomena – nationalism; sustainability/corporate social responsibility;
18
19
government action via public-private partnerships; and “grand challenges”
20
21
22 The rise of nationalism. In recent times, several parts in the world have experienced a
23
24 surge in nationalism (Financial Times, 2019). Research in management has remained mostly silent
25
26 on this important phenomenon (Ertug et al., 2018). However, insights from decades of work in
27
28
29 political science and social psychology (e.g., Mead, 1929; Druckman, 1994; Davidov, 2009) have
30
31 shown that nationalism is linked to behaviors that are likely to be relevant for research in
32
33 management, organizations, and strategy. For example, and as particularly relevant for TCT and
34
35
36
governance choices, higher levels of nationalism have been linked to a lower tendency to view
37
38 foreigners as being trustworthy (e.g., Yzerbyt & Demoulin, 2010), and to favoritism toward
39
40 domestic actors (e.g., Shi & Wright, 2003). Among other matters, these might have implications for
41
42
global supply chains and buyer-supplier relationships between firms from different countries. A
43
44
45 buyer from a more nationalistic country, as compared to one from a less nationalistic country, might
46
47 have a tendency to view foreign suppliers as less trustworthy. As a result, a buyer from a more
48
49 nationalistic country might perceive a greater threat of opportunistic behavior from their partners,
50
51
52 and therefore be more likely to try to mitigate this by internalizing the transaction (i.e. make rather
53
54 than buy), rather than to exchange with a foreign supplier. Nationalism might also matter on the
55
56
57
58 58
59
60
Page 59 of 90 Academy of Management Annals

1
2
3 supplier’s side in this relationship. As we mentioned, nationalism has been linked to a tendency to
4
5
6 favor domestic actors over foreign ones. Hence, a buyer might be more concerned that a foreign
7
8 supplier from a more nationalistic country will be more likely to behave opportunistically, in the
9
10 sense of favoring domestic firms/buyers at the expense of the focal buyer, than when dealing with
11
12
13
a foreign supplier from a less nationalistic country. This suggests that a buyer might be more likely
14
15 to internalize a transaction when a supplier is from a more nationalistic country. These two examples
16
17 point to the opportunities to investigate the impact of nationalism on the strategic and governance
18
19
decisions of firms from a TCT perspective.
20
21
22 Corporate social responsibility. Corporate social responsibility (CSR) has received
23
24 dramatically increased attention in recent years. Key issues in this stream of work include the
25
26 extent to which ostensible commitment to CSR is reflected in actual change, whether engagement
27
28
29 in CSR affects economic performance or competitive advantage, the effective monitoring of firms’
30
31 CSR promises, and the relative efficacy of different actors in undertaking socially beneficial
32
33 efforts. TCT has direct implications for all of these questions.
34
35
36
In a series of papers, Flammer and colleagues provide evidence that strategic investment
37
38 in CSR can yield bottom-line benefits to firms. The thread through many of these studies is that a
39
40 firm’s CSR investment can credibly signal its intention to behave responsibly in the future, in part
41
42
because its investment in a reputation for responsibility will be eroded otherwise (Flammer, 2020).
43
44
45 Such investments can also create favorable impressions on employees, customers, or bestowers of
46
47 government contracts, and these give the firm an advantage in the face of competitive pressures
48
49 (Flammer & Luo, 2017; Flammer, 2015). Relatedly, Henisz, Dorobantu, and Nartey (2014) find
50
51
52 that judicious stakeholder engagement enhances the market value of gold mining firms, even if
53
54 their underlying assets remain unchanged. In a cross-national study, Ioannis and Serafeim (2012)
55
56
57
58 59
59
60
 Academy of Management Annals Page 60 of 90

1
2
3 find that national institutions affect firms’ likelihood of engaging in CSR, presumably because of
4
5
6 differences in the expected returns to these actions.
7
8 Other scholars explore the degree to which firms may decouple their actual operations from
9
10 their professed claims to pursue CSR, and the ways in which such claims can be monitored (e.g.,
11
12
13
(e.g., Cuypers, Koh, & Wang, 2016; Luo, Wang, & Zhang, 2017). Of particular note here is the
14
15 fact that many CSR initiatives require concomitant efforts by other actors throughout a supply
16
17 chain, which makes monitoring and enforcement both crucial and challenging – and which fits
18
19
especially well with the TCT agenda. Toffel and colleagues have provided intriguing evidence
20
21
22 regarding the challenge of monitoring suppliers’ CSR-related practices, which can have substantial
23
24 reputational spillovers for a corporation. Short, Toffel, and Hugill (2016) find that privately
25
26 contracted “social auditors” are less likely to report problems when they are paid by the supplier,
27
28
29 have a prior relationship with the supplier, or are less well trained. Bird, Short, and Toffel (2019)
30
31 similarly find that incentive structures (both within and outside the supplier) affect the extent to
32
33 which suppliers truly adopt codes of conduct. Finally, in a within-firm setting, Pierce and Toffel
34
35
36
(2013) demonstrate that organization scope and structure directly influence internal monitoring of
37
38 unethical behavior.
39
40 These studies point to yet more fascinating questions illustrating how TCT can be fruitfully
41
42
applied to CSR research. Given the feasibility of public, private, or NGO monitoring of CSR
43
44
45 efforts, which form should be used for which type of CSR activity? Analogous to research on
46
47 industry-wide vs. firm-specific lobbying (e.g., de Figueiredo & Tiller, 2001; Jia, 2017), when are
48
49 industry-wide certification efforts preferable to firm-specific efforts? And, what are the tradeoffs
50
51
52 in encouraging CSR behavior? Flammer, Hong, and Minor (2019) find that, not surprisingly,
53
54 making CSR a formal part of the CEO’s compensation criteria yields greater corporate attention
55
56
57
58 60
59
60
Page 61 of 90 Academy of Management Annals

1
2
3 to CSR. However, is this costless? As the CEO devotes more effort to CSR, what other aspects of
4
5
6 the firm’s strategy get less attention and resources, and how does this affect firm performance? As
7
8 agency theory tells us, the wider the range of performance targets managers must meet, the more
9
10 difficult it is to monitor their performance (Williamson, 1963; Jensen & Meckling, 1976).
11
12
13
Therefore, what are the implications of moving to a more expansive set of performance criteria,
14
15 which now emcompass CSR?
16
17 Finally, two papers by Kaul and Luo distinguish between the private and social benefits of
18
19
CSR through a comparative governance lens. Kaul and Luo (2018) explore for-profit and non-
20
21
22 profit provision of social goods, demonstrating conditions under which a for-profit-firm’s CSR
23
24 effort will generate private returns but no social benefit. They develop a “comparative efficiency”
25
26 typology that matches organization form to CSR activity, and expand this in Luo and Kaul (2019)
27
28
29 to consider the comparative efficiency of a wider set of market frictions and organization types.
30
31 Drawing directly on ideas of discrete comparative analysis (Williamson, 1991), these studies
32
33 introduce TCT to a new research field and lay out a roadmap to address key questions about the
34
35
36
governance of CSR efforts.
37
38 Public-Private Partnerships. There has also been an upsurge of interest in public-private
39
40 partnerships (PPPs), as vehicles through which government can address important public-good
41
42
problems (Mahoney, McGahan, & Pitelis, 2009). We find TCT to be well-suited for contributing
43
44
45 to further addressing important questions in this area as well (Rangan, Samii, & Van Wassenhove,
46
47 2006; Kivleniece & Quelin, 2012; Chong, Saussier, & Silverman, 2015), just as it has been
48
49 fundamental to understanding collaboration between private firms (Hennart, 1988; Oxley, 1999;
50
51
52 Sampson, 2004). There are two particularly interesting aspects to PPPs from a TCT perspective.
53
54 First, public contracting is complicated because out-of-office political parties have an incentive to
55
56
57
58 61
59
60
 Academy of Management Annals Page 62 of 90

1
2
3 accuse the in-office party of exercising poor judgment or excessive discretion in these contracts.
4
5
6 Consequently, public-private contracts tend to be more rigid than private contracts (Moszoro et
7
8 al., 2016; Beuve et al., 2019). This rigidity constrains the value that can be created by PPPs.
9
10 Second, whereas TCT typically assumes that within-organization governance is more flexible than
11
12
13
inter-organizational governance (because of the possibility of exercising managerial fiat), in the
14
15 public sector this is frequently not the case – civil service rules often constrain in-house activity
16
17 more than public contracting will. Thus, the costs and benefits of PPPs are qualitatively different
18
19
from those of private-sector collaborations. Not only can TCT continue to contribute to our
20
21
22 understanding of PPPs, but a systematic study of public-private contracting can also serve to refine
23
24 the boundary conditions of TCT.
25
26 Grand challenges. Yet another set of phenomena that has gained importance in recent
27
28
29 years is related to what are referred to as “grand challenges” (e.g., George, Howard-Grenville,
30
31 Joshi, & Tihanyi, 2016). Addressing these grand challenges often requires collaboration between
32
33 diverse types of organizations (e.g., for profit firms, not-for profit firms, governments), which
34
35
36
might require different ways of organizing than typical collaborations. Hence, we see potential for
37
38 TCT to inform the managers and decision makers, both of firms and of other types of organizations,
39
40 about how these more diverse and complex collaborations should be governed.
41
42
In Table 4, we provide an overview of our discussion in this section.
43
44
45 <<< INSERT TABLE 4 HERE >>>
46
47 CONCLUSION
48
49 Transaction cost theory has enjoyed substantial influence across a wide swath of
50
51
52 management research and cognate disciplines. The theory has proven resilient due to its ability to
53
54 evolve effectively – to extend to new phenomena and to incorporate a broader set of theoretically
55
56
57
58 62
59
60
Page 63 of 90 Academy of Management Annals

1
2
3 relevant factors, while retaining a consistent perspective on the primary motivations for economic
4
5
6 organization. Yet, as a theory progresses, especially across a range of fields and phenomena, it is
7
8 important to intermittently devote attention to consolidating and integrating its advances and
9
10 challenges. This review is our attempt to provide such a consolidation, and to suggest a roadmap for
11
12
13
future research – notably, encouraging greater engagement with recent advances in relevant cognate
14
15 disciplines and highlighting how TCT can be applied to some of the most important novel
16
17 phenomena of our time. In sum, in addition to its proud history, we are confident that TCT
18
19
scholarship has a promising future awaiting.
20
21
22
23
24
25
26 REFERENCES
27
28
29 Abowd, J.A., & Lemieux, T. 1993. The effects of product market competition on collective
30 bargaining agreements: The case of foreign competition in Canada. The Quarterly Journal of
31 Economics, 108(4), 983-1014.
32
33 Adner, R., & Kapoor, R. 2010. Value creation in innovation ecosystems: How the structure of
34
technological interdependence affects firm performance in new technology generations. Strategic
35
36
Management Journal, 31(3), 306-333.
37
38 Ahmadjian, C., & Oxley, J.E. 2005. Using hostages to support exchange: Dependence balancing
39 and partial equity stakes in Japanese automotive supply relationships. Journal of Law,
40 Economics, and Organization, 22(1), 213-233.
41
42
Amit, R., & Zott, C. 2001. Value creation in e‐business. Strategic Management Journal, 22(6‐7),
43
44 493-520.
45
46 Anderson, E., & Gatignon, H. 1986. Modes of foreign entry: A transaction cost analysis and
47 propositions. Journal of International Business Studies, 17(3), 1-26.
48
49 Anderson, E., & Schmittlein, D.C. 1984. Integration of the sales force: An empirical examination.
50
51
The Rand Journal of Economics, 385-395.
52
53 Anderson, E., & Weitz, B. 1992. The use of pledges to build and sustain commitment in
54 distribution channels. Journal of Marketing Research, 29(1), 18-34.
55
56
57
58 63
59
60
 Academy of Management Annals Page 64 of 90

1
2
3 Anderson, J.C., & Narus, J.A. 1990. A model of distributor firm and manufacturer firm working
4
5
partnerships. Journal of Marketing, 54(1), 42-58.
6
7 Anderson, S.W., Dekker, H.C., & Van den Abbeele, A. 2017. Costly control: An examination of
8 the trade-off between control investments and residual risk in interfirm
9 transactions. Management Science, 63(7), 2163-2180.
10
11
Argyres, N.S., Bercovitz, J., & Mayer, K.J. 2007. Complementarity and evolution of contractual
12
13
provisions: An empirical study of IT services contracts. Organization Science, 18(1), 3-19.
14
15 Argyres, N.S., & Bigelow, L. 2007. Does transaction misalignment matter for firm survival at all
16 stages of the industry life cycle? Management Science, 53(8), 1332-1344.
17
18 Argyres, N.S., Felin, T., Foss, N., & Zenger, T. 2012. Organizational economics of capability and
19
heterogeneity. Organization Science, 23(5), 1213-1226.
20
21
22 Argyres, N.S., & Liebeskind, J.P. 1999. Contractual commitments, bargaining power, and
23 governance inseparability: Incorporating history into transaction cost theory. Academy of
24 Management Review, 24(1), 49-63.
25
26 Argyres, N.S., Mahoney, J.T., & Nickerson, J.A. 2019. Strategic responses to shocks: Comparative
27
28
adjustment costs, transaction costs, and opportunity costs. Strategic Management Journal,
29 40(3): 357-376.
30
31 Argyres, N.S., & Mayer, K.J. 2007. Contract design as a firm capability: An integration of learning
32 and transaction cost perspectives. Academy of Management Review, 32(4), 1060-1077.
33
34
Argyres, N.S., Rios, L.A., & and Silverman, B.S. 2020. Organizational change and the dynamics
35
36
of innovation: Formal R&D structure and intrafirm inventor networks. Strategic Management
37 Journal, forthcoming.
38
39 Argyres, N.S., & Silverman, B.S. 2004. R&D, organization structure, and the development of
40 corporate technological knowledge. Strategic Management Journal, 25(8-9), 929-958.
41
42
Ariely D. 2008. Predictably irrational. New York: HarperCollins.
43
44
45 Arora, A., & Gambardella, A. 2010. The market for technology. In Handbook of the Economics
46 of Innovation (Vol. 1, pp. 641-678). North-Holland.
47
48 Arrow, K.J. 1962. Economic Welfare and the Allocation of Resources for Invention. In K. Arrow
49 & R. Nelson (Ed.), The Rate and Direction of Inventive Activity. Princeton: Princeton University
50
51
Press, pp. 609-625.
52
53 Arrow, K.J. 1971. Insurance, risk and resource allocation. Essays in the Theory of Risk-bearing,
54 134-143.
55
56
57
58 64
59
60
Page 65 of 90 Academy of Management Annals

1
2
3 Arrow, K.J. 1974. The limits of organization. WW Norton & Company.
4
5
6 Arrow, K.J. 1984. The economics of information (Vol. 4). Harvard University Press.
7
8 Arthur, W.B. 1994. Increasing returns and path dependence in the economy. University of
9 Michigan Press, Ann Arbor, Michigan.
10
11
Betts, K.D., & Jaep, K.R. 2017. The Dawn of Fully Automated Contract Drafting: Machine
12
13
Learning Breathes New Life Into a Decades-Old Promise. Duke Law & Technology Review, 15,
14 216.
15
16 Beugelsdijk, S., Kostova, T., Kunst, V.E., Spadafora, E., & van Essen, M. 2018. Cultural distance
17 and firm internationalization: A meta-analytical review and theoretical implications. Journal of
18 Management, 44(1), 89-130.
19
20
21 Beuve, J., Moszoro, M.W., & Saussier, S. 2019. Political contestability and public contract
22 rigidity: An analysis of procurement contracts. Journal of Economics & Management
23 Strategy, 28(2), 316-335.
24
25 Bird, Y., Short, J.L., & Toffel, M.W. 2019. Coupling Labor Codes of Conduct and Supplier Labor
26 Practices: The Role of Internal Structural Conditions. Organization Science, 30(4), 847-867.
27
28
29 Boersma, M.F., Buckley, P.J., & Ghauri, P.N. 2003. Trust in international joint venture
30 relationships. Journal of Business Research, 56(12), 1031-1042.
31
32 Boudreau, K. 2010. Open platform strategies and innovation: Granting access vs. devolving
33 control. Management Science, 56(10), 1849-1872.
34
35
36
Boudreau, K. 2017. Platform-Based Organization and Boundary Choices: Entrepreneurship,
37 Innovation and Growth in the Early History of Mobile Computing. In Advances in Strategic
38 Management, volume 37. Eds. J. Furman, J. Gawer, A., Stern, S., Silverman, B.S. Emerald
39 Press.
40
41 Brahm, F. & Tarzijan, J. 2014. Transactional hazards, institutional change, and capabilities:
42
Integrating the theories of the firm. Strategic Management Journal, 35(2): 224-245.
43
44
45 Bromiley, P. & Cummings, L.L. 1995 Transaction Costs in Organizations with Trust. In
46 Research on Negotiation in Organizations. Eds. Bies, R., Sheppard, B. Lewicki, R.
47 Greenwich, CT: JAI Press.
48
49 Brouthers, K. & Hennart, J.F. 2007. Boundaries of the firm: insights from international entry mode
50
51
research. Journal of Management, 33(3), 395-425.
52
53 Bruce, J.R., de Figueiredo, J.M., & Silverman, B.S. 2019. Public contracting for private
54 innovation: Government capabilities, decision rights, and performance outcomes. Strategic
55 Management Journal, 40(4), 533-555.
56
57
58 65
59
60
 Academy of Management Annals Page 66 of 90

1
2
3
4
5
Buckley, P.J., & Casson, M. 1976. The Future of the Multinational Enterprise. New York:
6 McMillan Company
7
8 Buckley, P.J., & Chapman, M. 1998. The perception and measurement of transaction costs. In
9 International Business (pp. 57-86). Palgrave Macmillan, London.
10
11
Carson, S.J., Madhok, A., & Wu, T. 2006. Uncertainty, opportunism, and governance: The effects
12
13
of volatility and ambiguity on formal and relational contracting. Academy of Management
14 Journal, 49(5), 1058-1077.
15
16 Catalini, C., & Gans, J.S. 2016. Some simple economics of the blockchain (No. w22952).
17 National Bureau of Economic Research.
18
19
Chiles, T.H., & McMackin, J.F. 1996. Integrating variable risk preferences, trust, and transaction
20
21 cost economics. Academy of Management Review, 21(1), 73-99.
22
23 Coase, R.H. 1937. The nature of the firm. Economica, 4(16), 386-405.
24
25 Chong, E., Saussier, S., & Silverman B.S. 2015. Water under the bridge: Determinants of franchise
26 renewal in water provision. Journal of Law, Economics, and Organization, 31(1): 3-39.
27
28
29 Connelly, B.L., Certo, S.T., Ireland, R.D., & Reutzel, C.R. 2011. Signaling theory: A review and
30 assessment. Journal of Management, 37(1), 39-6
31
32 Conner, K.R., & Prahalad, C.K. 1996. A resource-based theory of the firm: Knowledge versus
33 opportunism. Organization Science, 7(5), 477-501.
34
35
36
Contractor, F., Kundu, S. & Hsu, C. 2003. A three-stage theory of international expansion: the link
37 between multinationality and performance in the service sector, Journal of International
38 Business Studies, 34(1): 5-18.
39
40 Cuypers, I.R., & Martin, X. 2007. Joint ventures and real options: An integrated perspective.
41 Advances in Strategic Management, 24: 107-148.
42
43
44 Cuypers, I. R., & Martin, X. 2010. What makes and what does not make a real option? A study of
45 equity shares in international joint ventures. Journal of International Business Studies, 41(1),
46 47-69.
47
48 Cuypers, I.R., Ertug, G., & Hennart, J.F. 2015. The effects of linguistic distance and lingua franca
49 proficiency on the stake taken by acquirers in cross-border acquisitions. Journal of International
50
51
Business Studies, 46(4), 429-442.
52
53 Cuypers, I.R., Koh, P.S., & Wang, H. 2016. Sincerity in corporate philanthropy, stakeholder
54 perceptions and firm value. Organization Science, 27(1), 173-188.
55
56
57
58 66
59
60
Page 67 of 90 Academy of Management Annals

1
2
3 David, R.J., & Han, S.K. 2004. A systematic assessment of the empirical support for transaction
4
5
cost economics. Strategic Management Journal, 25(1), 39-58.
6
7 Davidov, E. 2009. Measurement equivalence of nationalism and constructive patriotism in the
8 ISSP: 34 countries in a comparative perspective. Political Analysis, 17(1): 64-82.
9
10 Davidson, S., De Filippi, P., & Potts, J. 2018. Blockchains and the economic institutions of
11
capitalism. Journal of Institutional Economics, 14(4), 639-658.
12
13
14 Davidson, W. & McFetridge, D. 1984. International technology transactions and the theory of the
15 firm. Journal of Industrial Economics 32(3), 253-64.
16
17 de Figueiredo, J.M., & Silverman, B.S. 2012. Firm survival and industry evolution in vertically
18 related populations. Management Science, 58(9): 1632-1650.
19
20
21 de Figueiredo, J.M., & Tiller, E.H. 2001. The structure and conduct of corporate lobbying: How
22 firms lobby the Federal Communications Commission. Journal of Economics & Management
23 Strategy, 10(1), 91-122.
24
25 Druckman, D. 1994. Nationalism, patriotism, and group loyalty: A social psychological
26 perspective. Mershon International Studies Review, 38 (Supplement 1): 43-68.
27
28
29 Dunning, J., & Lundan, S. 2008. Multinational enterprises and the global economy. Cheltenham:
30 Edward Elgar.
31
32 Dyer, J.H., & Chu, W. 2003. The role of trustworthiness in reducing transaction costs and
33 improving performance: Empirical evidence from the United States, Japan, and Korea.
34
Organization Science, 14(1), 57-68.
35
36
37 Dyer, J.H., & Singh, H. 1998. The relational view: Cooperative strategy and sources of
38 interorganizational competitive advantage. Academy of Management Review, 23(4), 660-679.
39
40 Ertug, G., Cuypers, I., & Dow, D. 2018. Bringing Nationalism into Management Research:
41 Governance Choices in Cross-Border Collaborations. Academy of Management Proceedings, 1:
42
11967.
43
44
45 Ertug, G., Gargiulo, M., Galunic, C., & Zou, T. 2018. Homophily and individual performance.
46 Organization Science, 29(5), 912-930.
47
48 Felin, T., Foss, N.J., & Ployhart, R.E. 2015. The microfoundations movement in strategy and
49 organization theory. The Academy of Management Annals, 9(1), 575-632.
50
51
52 Feltchenhauer, D., & Dunning, D. 2010. Why so cynical? Asymmetric feedback underlies misguided
53 skepticism regarding the trustworthiness of others. Psychological Science, 21(2): 189-193.
54
55 Financial Times. 2019. The global advance of ethno-nationalism. (Edward Luce, 24 May, 2019).
56
57
58 67
59
60
 Academy of Management Annals Page 68 of 90

1
2
3
4
5
Fladmoe-Lindquist, K., & Jacque, L.L. 1995. Control modes in international service operations:
6 The propensity to franchise. Management Science, 41(7), 1238-1249.
7
8 Fladmoe-Lindquist, K. & Jacque, L.L. 2005. To own or to franchise: International control
9 decisions for service companies. Journal of Applied Corporate Finance 9(3), 98-109.
10
11
Flammer, C. 2015. Does corporate social responsibility lead to superior financial performance? A
12
13
regression discontinuity approach. Management Science, 61(11), 2549-2568.
14
15 Flammer, C. 2020. Corporate green bonds. Journal of Financial Economics. Forthcoming.
16
17 Flammer, C., Hong, B., & Minor, D. 2019. Corporate governance and the rise of integrating
18 corporate social responsibility criteria in executive compensation: Effectiveness and implications
19
for firm outcomes. Strategic Management Journal, 40(7), 1097-1122.
20
21
22 Flammer, C., & Luo, J. 2017. Corporate social responsibility as an employee governance tool:
23 Evidence from a quasi‐experiment. Strategic Management Journal, 38(2), 163-183.
24
25 Foss, N.J. 1996a. Knowledge-based approaches to the theory of the firm: Some critical comments.
26 Organization Science, 7(5), 470-476.
27
28
29 Foss, N.J. 1996b. More critical comments on knowledge-based theories of the firm. Organization
30 Science, 7(5), 519-523.
31
32 Forbes, S.J., & Lederman, M. 2010. Does vertical integration affect firm performance? Evidence
33 from the airline industry. The RAND Journal of Economics, 41(4), 765-790.
34
35
36
French, M.T., & Popovici, I. 2011. That instrument is lousy! In search of agreement when using
37 instrumental variables estimation in substance use research. Health Economics, 20(2), 127-146.
38
39 Ganesan, S. 1994. Determinants of long-term orientation in buyer-seller relationships. Journal of
40 Marketing, 58(2), 1-19.
41
42
Gans, J.S., & Stern, S. 2010. Is there a market for ideas?. Industrial and Corporate Change, 19(3),
43
44 805-837.
45
46 Gargiulo, M. & Ertug, G. 2006. The dark side of trust. In R. Bachmann, R. & Zaheer, A. (Eds.),
47 Handbook of Trust Research. Edward Elgar.
48
49 Garrett, T.A. 2003. Aggregated versus disaggregated data in regression analysis: implications for
50
51
inference. Economics Letters, 81(1), 61-65.
52
53 Gartenberg, C., & Pierce, L. 2017. Subprime governance: Agency costs in vertically integrated
54 banks and the 2008 mortgage crisis. Strategic Management Journal, 38(2), 300-321.
55
56
57
58 68
59
60
Page 69 of 90 Academy of Management Annals

1
2
3 Gatignon, H., & Anderson, E. 1988. The multinational corporation's degree of control over foreign
4
5
subsidiaries: An empirical test of a transaction cost explanation. Journal of Law, Economics &
6 Organization, 4, 305.
7
8 George, G., Howard-Grenville, J., Joshi, A., & Tihanyi, L. 2016. Understanding and tackling
9 societal grand challenges through management research. Academy of Management Journal,
10 59(6), 1880-1895.
11
12
13
Geyskens, I., Steenkamp, J.B.E., & Kumar, N. 2006. Make, buy, or ally: A transaction cost theory
14 meta-analysis. Academy of Management Journal, 49(3), 519-543.
15
16 Ghoshal, S. 2005. Bad management theories are destroying good management practices. Academy
17 of Management Learning & Education, 4(1), 75-91.
18
19
Ghoshal, S., & Moran, P. 1996. Bad for practice: A critique of the transaction cost theory.
20
21 Academy of Management Review, 21(1), 13-47.
22
23 Gibbons, R., Holden, R., & Powell, M. 2012. Organization and information: firms’ governance
24 choices in rational-expectations equilibrium. The Quarterly Journal of Economics, 127(4),
25 1813-1841.
26
27
28
Gigerenzer, G., & Gaissmaier, W. 2011. Heuristic decision making. Annual Review of
29 Psychology, 62, 451-482.
30
31 Goffman, E. 1969. Strategic Interaction. Philadelphia: University of Pennsylvania Press.
32
33 Gomes-Casseres, B. 2003. Competitive advantage in alliance constellations. Strategic
34
Organization, 1(3), 327-335.
35
36
37 Graebner, M.E. 2009. Caveat venditor: Trust asymmetries in acquisitions of entrepreneurial firms.
38 Academy of Management Journal, 52(3), 435-472.
39
40 Granovetter, M. 1985. Economic action and social structure: The problem of embeddedness.
41 American Journal of Sociology, 91(3), 481-510.
42
43
44 Grossman, S.J., & Hart, O.D. 1986. The costs and benefits of ownership: A theory of vertical and
45 lateral integration. Journal of Political Economy, 94(4), 691-719.
46
47 Guertzgen, N. 2009. Rent-sharing and collective bargaining coverage: Evidence from linked
48 employer-employee data. Scandinavian Journal of Economics, 111(2): 323-349.
49
50
51
Gulati, R. 1995. Does familiarity breed trust? The implications of repeated ties for contractual
52 choice in alliances. Academy of Management Journal, 38(1), 85-112.
53
54 Gulati, R., Lavie, D., & Singh, H. 2009. The nature of partnering experience and the gains from
55 alliances. Strategic Management Journal, 30(11), 1213-1233.
56
57
58 69
59
60
 Academy of Management Annals Page 70 of 90

1
2
3
4
5
Gulati, R., & Nickerson, J.A. 2008. Interorganizational trust, governance choice, and exchange
6 performance. Organization Science, 19(5), 688-708.
7
8 Hagiu, A. 2014. Strategic decisions for multisided platforms. Sloan Management Review, 55(2):
9 71-80.
10
11
Hagiu, A., & Wright, J. 2015. Multi-sided platforms. International Journal of Industrial
12
13
Organization, 43, 162-174.
14
15 Haleblian, J., & Finkelstein, S. 1999. The influence of organizational acquisition experience on
16 acquisition performance: A behavioral learning perspective. Administrative Science Quarterly,
17 44(1), 29-56.
18
19
Hamilton, B.A. & Nickerson, J.A. 2003. Correcting for Endogeneity in Strategic Management
20
21 Research. Strategic Organization, 1(1): 51-78.
22
23 Harmon, D.J., Kim, P.H., & Mayer, K.J. 2015. Breaking the letter vs. spirit of the law: How the
24 interpretation of contract violations affects trust and the management of relationships. Strategic
25 Management Journal, 36(4), 497-517.
26
27
28
Hart, O., & Moore, J. 1990. Property Rights and the Nature of the Firm. Journal of Political
29 Economy, 98(6), 1119-1158.
30
31 Heide, J.B., & John, G. 1988. The role of dependence balancing in safeguarding transaction-
32 specific assets in conventional channels. Journal of Marketing, 52(1), 20-35.
33
34
Henisz, W.J. 2000. The institutional environment for multinational investment. Journal of Law,
35
36
Economics, and Organization, 16(2), 334-364.
37
38 Henisz, W.J., Dorobantu, S., & Nartey, L. 2014. Spinning gold: The financial returns to
39 stakeholder engagement. Strategic Management Journal, 35(12), 1727-1748.
40
41 Henisz, W.J., & Macher, J.T. 2004. Firm- and country-level tradeoffs and contingencies in the
42
evaluation of foreign investment: The semiconductor industry, 1994-2002. Organization Science,
43
44 15(5), 537-554.
45
46 Hennart, J.-F. 1982. A Theory of Multinational Enterprise. Ann Arbor: University of Michigan
47 Press.
48
49 Hennart, J.-F. 1988a. Upstream vertical integration in the World aluminum and tin industries: a
50
51
comparative study of the choice between market and intrafirm coordination. Journal of
52 Economic Behavior and Organization, 9(3), 281-300.
53
54 Hennart, J.-F. 1988b. A transaction costs theory of equity joint ventures. Strategic Management
55 Journal, 9(4), 361-374.
56
57
58 70
59
60
Page 71 of 90 Academy of Management Annals

1
2
3
4
5
Hennart, J.-F. 1989. The transaction cost rationale for countertrade. Journal of Law, Economics,
6 and Organization, 5(1)1, 127-153.
7
8 Hennart, J.-F. 1991. Control in multinational firms: The role of price and hierarchy. Management
9 International Review, 30th anniversary issue, 71-96.
10
11
Hennart, J.-F. 1993. Explaining the Swollen Middle: Why most transactions are a mix of market
12
13
and hierarchy. Organization Science, 4(4): 529-548.
14
15 Hennart, J.-F. 2007. The theoretical rationale for a multinationality-performance relationship.
16 Management International Review, 47(3), 423-452.
17
18 Hennart, J.-F. 2009. Down with MNE-centric theories! Market Entry and Expansion as the
19
Bundling of MNE and Local Assets, Journal of International Business Studies, 40(9), pp. 1432-
20
21 1454.
22
23 Hennart, J.-F. 2011. A theoretical assessment of the empirical literature on the impact of
24 multinationality on performance. Global Strategy Journal, 1(1‐2), 135-151.
25
26 Hennart, J.-F. 2013. Internal and external hybrids and the nature of joint venture. In A. Grandori
27
28
(Ed.), Handbook of Economic Organization: Integrating Economic and Organization Theory.
29 Cheltenham: Edward Elgar, pp. 481-500.
30
31 Hennart, J.-F. 2015a. The transaction cost theory of the Transnational Corporation” in Sarianna
32 Lundan, (ed.) Transnational Corporations and Transnational Governance. Basingstoke:
33 Palgrave Macmillan, pp. 25-52.
34
35
36 Hennart, J.-F. 2015b. Leveraging Asian institutions to deepen theory: A transaction cost
37 perspective on relational governance. Asian Business and Management, 14(4), 257-282.
38
39 Hennart, J.-F. 2019. Digitalized service multinationals and international business theory. Journal
40 of International Business Studies, 50(8), 1388-1400.
41
42
Hennart, J.-F. & Park, Y.R. 1993. Greenfield vs. acquisition: The strategy of Japanese investors in
43
44 the United States. Management Science, 39(9), 1054-1070.
45
46 Hennart, J.-F. & Park, Y.R. 1994. Location, governance, and strategic determinants of Japanese
47 manufacturing investment in the United States. Strategic Management Journal 15(6), 419-436.
48
49 Hoetker, G., & Mellewigt, T. 2009. Choice and performance of governance mechanisms: matching
50
51
alliance governance to asset type. Strategic Management Journal, 30(10), 1025-1044.
52
53 Holmstrom, B., & Milgrom, P. 1991. Multitask principal-agent analyses: Incentive contracts, asset
54 ownership, and job design. Journal of Law, Economics & Organization, 7, 24-52.
55
56
57
58 71
59
60
 Academy of Management Annals Page 72 of 90

1
2
3 Ioannou, I., & Serafeim, G. 2012. What drives corporate social performance? The role of nation-
4
5
level institutions. Journal of International Business Studies, 43(9), 834-864.
6
7 Jap, S. D., & Anderson, E. 2003. Safeguarding interorganizational performance and continuity
8 under ex post opportunism. Management Science, 49(12), 1684-1701.
9
10 Jarrahi, M.H. 2018. Artificial intelligence and the future of work: human-AI symbiosis in
11
organizational decision making. Business Horizons, 61(4), 577-586.
12
13
14 Jensen, M.C., & Meckling, W.H. 1976. Theory of the firm: Managerial behavior, agency costs and
15 ownership structure. Journal of Financial Economics, 3(4): 305-360.
16
17 Jia, N. 2018. The “make and/or buy” decisions of corporate political lobbying: Integrating the
18 economic efficiency and legitimacy perspectives. Academy of Management Review, 43(2), 307-
19
326.
20
21
22 Joskow, P.L. 1985. Vertical integration and long-term contracts: The case of coal-burning electric
23 generating plants. Journal of Law, Economics, & Organization, 1(1), 33-80.
24
25 Joskow, P.L. 1988. Asset specificity and the structure of vertical relationships: empirical evidence.
26 Journal of Law, Economics, & Organization, 4(1), 95-117.
27
28
29 Kahneman, D. 2011. Thinking, fast and slow. Macmillan.
30
31 Kahneman, D., Knetsch, J.L., & Thaler, R.H. 1990. Experimental tests of the endowment effect
32 and the Coase theorem. Journal of political Economy, 98(6), 1325-1348.
33
34
Kahneman, D., Slovic, S.P., Slovic, P., & Tversky, A. (Eds.). 1982. Judgment under uncertainty:
35
36
Heuristics and biases. Cambridge university.
37
38 Kano, L., & Verbeke, A. 2015. The three faces of bounded reliability: Alfred Chandler and the
39 micro-foundations of management theory. California Management Review, 58(1), 97-122.
40
41 Kaul, A., & Luo, J. 2018. An economic case for CSR: The comparative efficiency of for-profit
42
firms in meeting consumer demand for social goods. Strategic Management Journal, 39(6),
43
44 1650-1677.
45
46 Kellogg, K., Valentine, M., & Christin, A. 2019. Algorithms at work: The new contested terrain
47 of control. Academy of Management Annals, doi.org/10.5465/annals.2018.0174
48
49 Kenney, M., & Florida, R.L. 1993. Beyond Mass Production. Oxford University Press.
50
51
52 Ketokivi, M., & Mahoney, J.T. 2016. Transaction cost economics as a constructive stakeholder
53 theory. Academy of Management Learning & Education, 15(1), 123-138.
54
55
56
57
58 72
59
60
Page 73 of 90 Academy of Management Annals

1
2
3 Kivleniece, I., & Quelin, B.V. 2012. Creating and capturing value in public-private ties: A private
4
5
actor's perspective. Academy of Management Review, 37(2), 272-299.
6
7 Klein, B., Crawford, R.G., & Alchian, A.A. 1978. Vertical integration, appropriable rents, and the
8 competitive contracting process. The Journal of Law and Economics, 21(2), 297-326.
9
10 Klein, S. 1989. A transaction cost explanation of vertical control in international markets. Journal
11
of the Academy of Marketing Science, 17(3), 253-260.
12
13
14 Klein, S., Frazier, G.L., & Roth, V.J. 1990. A transaction cost analysis model of channel
15 integration in international markets. Journal of Marketing Research, 27(2), 196-208.
16
17 Koopmans, T.C. 1957. Three essays on the state of economic science. McGraw-Hill: New York.
18
19
Kosterman, R., & Feshbach, S. 1989. Toward a measure of patriotic and nationalistic attitudes.
20
21 Political psychology, 257-274.
22
23 Langfred, C.W. 2004. Too much of a good thing? Negative effects of high trust and individual
24 autonomy in self-managing teams. Academy of Management Journal, 47(3), 385-399.
25
26 Langlois, R.N. 1992. Transaction-cost economics in real time. Industrial and Corporate Change,
27
28
1(1), 99-127.
29
30 Lawrence, B.S. & Shah, N.P. 2020. Homophily: Measures and meaning. Academy of
31 Management Annals. doi.org/10.5465/annals.2018.0147
32
33 Lehdonvirta, V., Kässi, O., Hjorth, I., Barnard, H., & Graham, M. 2019. The global platform
34
economy: A new offshoring institution enabling emerging-economy microproviders. Journal of
35
36
Management, 45(2), 567-599.
37
38 Leiblein, M.J. 2003. The choice of organizational governance form and performance: Predictions
39 from transaction cost, resource-based, and real options theories. Journal of Management, 29(6),
40 937-961.
41
42
Leiblein, M.J., Reuer, J.J., & Dalsace, F. 2002. Do make or buy decisions matter? The influence
43
44 of organizational governance on technological performance. Strategic Management Journal,
45 23(9), 817-833.
46
47 Levin, R.C., Klevorick, A.K., Nelson, R.R., Winter, S.G. 1987. Appropriating the returns from
48 industrial research and development. Brookings Papers on Economic Activity, (3), 783-831.
49
50
51
Liu, L., & Weingast, B.R. 2017. Taobao, federalism, and the emergence of law, Chinese style.
52 Minnesota Law Review, 102(4): 1563-1590.
53
54 Loewenstein, G., Issacharoff, S., Camerer, C., & Babcock, L. 1993. Self-serving assessments of
55 fairness and pretrial bargaining. The Journal of Legal Studies, 22(1), 135-159.
56
57
58 73
59
60
 Academy of Management Annals Page 74 of 90

1
2
3
4
5
Lu, J. & Beamish, P. 2004. International diversification and firm performance: the S curve
6 hypothesis. Academy of Management Journal 47(4): 598-609.
7
8 Lu, J.W., & Hebert, L. 2005. Equity control and the survival of international joint ventures: a
9 contingency approach. Journal of Business Research, 58(6), 736-745.
10
11
Lui, S.S., & Ngo, H.Y. 2004. The role of trust and contractual safeguards on cooperation in non-
12
13
equity alliances. Journal of Management, 30(4), 471-485.
14
15 Luo, J., & Kaul, A. 2019. Private action in public interest: The comparative governance of social
16 issues. Strategic Management Journal, 40(4): 476-502.
17
18 Luo, X.R., Wang, D., & Zhang, J. 2017. Whose call to answer: Institutional complexity and firms’
19
CSR reporting. Academy of Management Journal, 60(1), 321-344.
20
21
22 Luo, Y. 2005. Transactional characteristics, institutional environment and joint venture contracts.
23 Journal of International Business Studies, 36(2), 209-230.
24
25 Macneil, I.R. 1974. The many futures of contracts. Southern California Law Review, 47, 691-
26 816.
27
28
29 Macneil, I.R. 1978. Contracts: Adjustment of long-term economic relations under classical,
30 neoclassical, and relational contract law. Northwestern University Law Review, 72, 854-906.
31
32 Mahoney, J.T., McGahan, A.M., & Pitelis, C.N. (2009). Perspective—The interdependence of
33 private and public interests. Organization Science, 20(6), 1034-1052.
34
35
36
Malhotra, D., & Lumineau, F. 2011. Trust and collaboration in the aftermath of conflict: The
37 effects of contract structure. Academy of Management Journal, 54(5), 981-998.
38
39 Maseland, R., Dow, D., & Steel, P. 2018. The Kogut and Singh national cultural distance index:
40 Time to start using it as a springboard rather than a crutch. Journal of International Business
41 Studies, 49(9), 1154-1166.
42
43
44 Masten, S.E. 1988. A legal basis for the firm. Journal of Law, Economics, & Organization, 4(1),
45 181-198.
46
47 Masten, S.E., Meehan Jr, J. W., & Snyder, E. A. 1991. The Costs of Organization. The Journal
48 of Law, Economics, and Organization, 7(1), 1-25.
49
50
51
Mayer, K.J., & Argyres, N.S. 2004. Learning to contract: Evidence from the personal computer
52 industry. Organization Science, 15(4), 394-410
53
54 Mayer, R.C., Davis, J.H., & Schoorman, F.D. 1995. An integrative model of organizational
55 trust. Academy of Management Review, 20(3), 709-734.
56
57
58 74
59
60
Page 75 of 90 Academy of Management Annals

1
2
3
4
5
McEvily, B. (2011). Reorganizing the boundaries of trust: From discrete alternatives to hybrid
6 forms. Organization Science, 22(5), 1266-1276.
7
8 McEvily, B., Perrone, V., & Zaheer, A. 2003. Trust as an organizing principle. Organization
9 Science, 14(1), 91-103.
10
11
McEvily, B., Soda, G., & Tortoriello, M. 2014. More formally: Rediscovering the missing link
12
13
between formal organization and informal social structure. Academy of Management
14 Annals, 8(1), 299-345.
15
16 McEvily, B., Zaheer, A., & Kamal, D.K.F. 2017. Mutual and exclusive: Dyadic sources of trust in
17 interorganizational exchange. Organization Science, 28(1), 74-92.
18
19
McQuade, K., & Gomes-Casseres, B. 1992. Xerox and Fuji Xerox. Harvard Business School case
20
21 9-391-156.
22
23 Mead, G.H. 1929. National-mindedness and International-mindedness. International Journal of
24 Ethics, 39: 385-407.
25
26 Mills, M. 2016. Artificial Intelligence in Law: The State of Play 2016. Thomson Reuters Legal
27
28
executive Institute.
29
30 Monteverde, K., & Teece, D.J. 1982. Supplier switching costs and vertical integration in the
31 automobile industry. The Bell Journal of Economics, 206-213.
32
33 Montgomery, C.A., & Hariharan, S. 1991. Diversified expansion by large established firms.
34
Journal of Economic Behavior & Organization, 15(1), 71-89.
35
36
37 Moszoro, M., Spiller, P.T., & Stolorz, S. 2016. Rigidity of public contracts. Journal of Empirical
38 Legal Studies, 13(3), 396-427.
39
40 Nickerson, J.A., & Silverman, B.S. 2003. Why firms want to organize efficiently and what keeps
41 them from doing so: Inappropriate governance, performance, and adaptation in a deregulated
42
industry. Administrative Science Quarterly, 48(3), 433-465.
43
44
45 Novak, S., & Stern, S. 2008. How does outsourcing affect performance dynamics? Evidence from
46 the automobile industry. Management Science, 54(12), 1963-1979.
47
48 Oxley, J.E. 1997. Appropriability Hazards and Governance in Strategic Alliances: A Transaction
49 Cost Approach. Journal of Law, Economics and Organization 13:387-409.
50
51
52 Oxley, J.E. 1999. Institutional environment and the mechanisms of governance: the impact of
53 intellectual property protection on the structure of inter-firm alliances. Journal of Economic
54 Behavior & Organization, 38(3), 283-309.
55
56
57
58 75
59
60
 Academy of Management Annals Page 76 of 90

1
2
3 Ouchi, W.G. 1981. Organizational paradigms: A commentary on Japanese management and
4
5
Theory Z organizations. Organizational Dynamics, 9(4), 36-43
6
7 Palay, T.M. 1985. Avoiding regulatory constraints: Contracting safeguards and the role of informal
8 agreements. Journal of Law, Economics, & Organization, 1(1), 155-175.
9
10 Pfeffer, J. 2005. Why do bad management theories persist? A comment on Ghoshal. Academy of
11
Management Learning & Education, 4(1), 96-100.
12
13
14 Perrow, C. 1986. Economic theories of organization. Theory and Society, 15(1-2), 11-45.
15
16 Pierce, L., & Toffel, M.W. 2013. The role of organizational scope and governance in strengthening
17 private monitoring. Organization Science, 24(5), 1558-1584.
18
19
Pilling, B.K., Crosby, L.A., & Jackson Jr, D.W. 1994. Relational bonds in industrial exchange: an
20
21 experimental test of the transaction cost economic framework. Journal of Business Research,
22 30(3), 237-251.
23
24 Pisano, G.P. 1990. The R&D boundaries of the firm: an empirical analysis. Administrative Science
25 Quarterly, 35(1), 153-176.
26
27
28
Poppo, L., & Zenger, T. 2002. Do formal contracts and relational governance function as
29 substitutes or complements?. Strategic Management Journal, 23(8), 707-725.
30
31 Poppo, L., Zhou, K.Z., & Li, J.J. 2016. When can you trust “trust”? Calculative trust, relational
32 trust, and supplier performance. Strategic Management Journal, 37(4), 724-741.
33
34
Powell, K.S. 2014. From M–P to MA–P: Multinationality alignment and performance. Journal of
35
36
International Business Studies, 45(2), 211-226.
37
38 Powell, W.W. 1987. Hybrid organizational arrangements: new form or transitional development?.
39 California Management Review, 30(1), 67-87.
40
41 Powell, W.W. 1990. Neither market nor hierarchy: Network forms of organization. Research in
42
Organizational Behavior, 12: 295-336.
43
44
45 Puranam, P., & Vanneste, B.S. 2009. Trust and governance: Untangling a tangled web. Academy
46 of Management Review, 34(1), 11-31.
47
48 Quintana, D. S., & Williams, D. R. 2018. Bayesian alternatives for common null-hypothesis
49 significance tests in psychiatry: a non-technical guide using JASP. BMC Psychiatry, 18(1), 178.
50
51
52 Rangan, S., Samii, R., & Van Wassenhove, L.N. 2006. Constructive partnerships: When alliances
53 between private firms and public actors can enable creative strategies. Academy of Management
54 Review, 31(3), 738-751.
55
56
57
58 76
59
60
Page 77 of 90 Academy of Management Annals

1
2
3 Reuer, J.J., & Ariño, A. 2002. Contractual renegotiations in strategic alliances. Journal of
4
5
Management, 28(1), 47-68.
6
7 Reuer, J.J., & Ariño, A. 2007. Strategic alliance contracts: Dimensions and determinants of
8 contractual complexity. Strategic Management Journal, 28(3), 313-330.
9
10 Reuer, J.J., & Devarakonda, S. V. 2016. Mechanisms of hybrid governance: Administrative
11
committees in non-equity alliances. Academy of Management Journal, 59(2), 510-533.
12
13
14 Reuer, J.J., Zollo, M., & Singh, H. 2002. Post‐formation dynamics in strategic alliances. Strategic
15 Management Journal, 23(2), 135-151.
16
17 Rich, B., Weber, L., & Bauman, C. 2020. Security and Optimism in Contracts in the Digital
18 Age. Research Handbook on Contract Design. Edward Elgar Publishing: New York.
19
20
21 Richman, B.D., & Macher, J.T. 2008. Transaction cost economics: An assessment of empirical
22 research in the social sciences. Business and Politics, 10, 1-63.
23
24 Rietveld, J., Seamans, R. & Meggiorin, K.. 2020. Market Orchestrators: The Effects of Platform
25 Certification on Complementor Behavior. Working paper, NYU.
26
27
28
Rindfleisch, A., & Heide, J.B. 1997. Transaction cost analysis: Past, present, and future
29 applications. Journal of Marketing, 61(4): 30–54.
30
31 Ring, P.S., & Van de Ven, A.H. 1994. Developmental processes of cooperative interorganizational
32 relationships. Academy of Management Review, 19(1), 90-118.
33
34
Robinson, D.T., & Stuart, T.E. 2007. Financial contracting in biotech strategic alliances. The
35
36 Journal of Law and Economics, 50(3), 559-596.
37
38 Rosenthal, R. 1979. The "file drawer problem" and tolerance for null results. Psychological
39 Bulletin, 86: 638-641.
40
41 Roth, A.E. 2008. What have we learned from market design? Economic Journal, 118:285-310.
42
43
44 Roth, A.E., & Murnighan, J.K. 1982. The role of information in bargaining: An experimental
45 study. Econometrica: Journal of the Econometric Society, 1123-1142.
46
47 Rousseau, D.M., Sitkin, S.B., Burt, R.S., & Camerer, C. 1998. Not so different after all: A cross-
48 discipline view of trust. Academy of Management Review, 23(3), 393-404.
49
50
51
Rugman, A.M. 1981. Inside the Multinationals, London. Croom Helm.
52
53 Sampson, R.C. 2004. The cost of misaligned governance in R&D alliances. Journal of Law,
54 Economics, and Organization, 20(2), 484-526.
55
56
57
58 77
59
60
 Academy of Management Annals Page 78 of 90

1
2
3 Schilke, O., & Cook, K.S. 2015. Sources of alliance partner trustworthiness: Integrating calculative
4
5
and relational perspectives. Strategic Management Journal, 36(2), 276-297.
6
7 Schilke, O., & Huang, L. 2018. Worthy of swift trust? How brief interpersonal contact affects trust
8 accuracy. Journal of Applied Psychology, 103(11), 1181.
9
10 Schmidt, C.G., & Wagner, S.M. 2019. Blockchain and supply chain relations: A transaction cost
11
theory perspective. Journal of Purchasing and Supply Management, 25(4), 100552.
12
13
14 Schneider, S., & Leyer, M. 2019. Me or information technology? Adoption of artificial intelligence
15 in the delegation of personal strategic decisions. Managerial and Decision Economics, 40(3),
16 223-231.
17
18 Schweitzer, M.E., Ho, T-H. & Zhang, X. 2018. How monitoring influences trust: A tale of two
19
faces. Management Science, 64(1), 253-270.
20
21
22 Shaver, J.M. 1998. Accounting for endogeneity when assessing strategy performance: Does entry
23 mode choice affect FDI survival?. Management Science, 44(4), 571-585.
24
25 Short, J.L., Toffel, M.W., & Hugill, A.R. (2016). Monitoring global supply chains. Strategic
26 Management Journal, 37(9), 1878-1897.
27
28
29 Shi, X., & Wright, P.C. 2003. The potential impacts of national feelings on international business
30 negotiations: a study in the China context. International Business Review, 12(3), 311-328.
31
32 Silverman, B.S. 1999. Technological resources and the direction of corporate diversification:
33 Toward an integration of the resource-based view and transaction cost economics. Management
34
Science, 45(8), 1109-1124.
35
36
37 Silverman, B.S. 2019. Markets for knowledge: Intellectual property, organizational arrangements,
38 and international governance. In Brousseau, E., J. Glachant & J. Sgard, (Eds.), Oxford Handbook
39 on International Economic Governance & Market Regulation. Oxford: Oxford University Press.
40
41 Simon, H.A. 1957. Models of man, social and rational. New York: Garland Publishing.
42
43
44 Spence, A. 1973. Job Market Signaling. The Quarterly Journal of Economics, 87(3), 355-374.
45
46 Spence, M. 2002. Signaling in retrospect and the informational structure of markets. American
47 Economic Review, 92(3), 434-459.
48
49 Standiford, S. & Marshall, R. 2000. The transaction cost advantage of guanxi-based business
50
51
practices. Journal of World Business, 35(1): 21–42.
52
53 Stinchcombe, A.L. 1990. Information and organizations (Vol. 19). University of California Press.
54
55
56
57
58 78
59
60
Page 79 of 90 Academy of Management Annals

1
2
3 Stuckey, J. 1983. Vertical Integration and Joint Ventures in the Aluminum Industry. Cambridge:
4
5
Harvard University Press.
6
7 Tambe, P., Cappelli, P., & Yakubovich, V. 2019. Artificial Intelligence in Human Resources
8 Management: Challenges and a Path Forward. California Management Review, 61(4), 15–42
9
10 Teece, D.J. 1985. Multinational enterprise, internal governance, and industrial organization. The
11
American Economic Review, 75(2), 233-238.
12
13
14 Teece, D.J. 1986. Profiting from technological innovation: Implications for integration,
15 collaboration, licensing and public policy. Research Policy, 15(6), 285-305.
16
17 Teece, D.J. 1986. Transactions cost economics and the multinational enterprise An Assessment.
18 Journal of Economic Behavior & Organization, 7(1), 21-45.
19
20
21 Teece, D.J. 2007. Explicating dynamic capabilities: the nature and microfoundations of
22 (sustainable) enterprise performance. Strategic Management Journal, 28(13), 1319-1350.
23
24 Tihanyi, L., Griffith, D.A., & Russell, C.J. 2005. The effect of cultural distance on entry mode
25 choice, international diversification, and MNE performance: A meta-analysis. Journal of
26 International Business Studies, 36(3), 270-283.
27
28
29 Tsang, E.W. 2006. Behavioral assumptions and theory development: The case of transaction cost
30 economics. Strategic Management Journal, 27(11), 999-1011.
31
32 Vanneste, B.S., Puranam, P., & Kretschmer, T. 2014. Trust over time in exchange relationships:
33 Meta‐analysis and theory. Strategic Management Journal, 35(12), 1891-1902.
34
35
36
Verbeke, A., & Greidanus, N. 2009. The end of the opportunism versus trust debate: Bounded
37 reliability as a new envelope concept in research on MNE governance. Journal of International
38 Business Studies, 40: 1472-1495.
39
40 Veugelers, R., & B. Cassiman. 1999. Make and buy in innovation strategies: Evidence from
41 Belgian manufacturing firms. Research Policy, 28(1): 63-80.
42
43
44 Walker, G., & Weber, D. 1984. A transaction cost approach to make-or-buy decisions.
45 Administrative Science Quarterly, 373-391.
46
47 Walker, G., & Weber, D. 1987. Supplier competition, uncertainty, and make-or-buy decisions.
48 Academy of Management Journal, 30(3), 589-596.
49
50
51
Weber, L., & Mayer, K. 2014. Transaction cost economics and the cognitive perspective:
52 Investigating the sources and governance of interpretive uncertainty. Academy of Management
53 Review, 39(3), 344-363.
54
55
56
57
58 79
59
60
 Academy of Management Annals Page 80 of 90

1
2
3 Williamson, O.E. 1963. Managerial discretion and business behavior. The American Economic
4
5
Review, 53(5), 1032-1057.
6
7 Williamson, O.E. 1973. Markets and hierarchies: some elementary considerations. The American
8 Economic Review, 63(2), 316-325
9
10 Williamson, O.E. 1975. Markets and Hierarchies: Analysis and Antitrust Implications. New
11
York: Free Press.
12
13
14 Williamson, O.E. 1979. Transaction-cost economics: the governance of contractual relations. The
15 journal of Law and Economics, 22(2), 233-261.
16
17 Williamson, O.E. 1981. The economics of organization: The transaction cost approach. American
18 Journal of Sociology, 87(3), 548-577.
19
20
21 Williamson, O.E. 1983. Credible commitments: Using hostages to support exchange. The
22 American Economic Review, 73(4), 519-540.
23
24 Williamson, O.E. 1985. The Economic Institutions of Capitalism. New York: Free Press.
25
26 Williamson, O.E. 1991. Comparative economic organization: The analysis of discrete structural
27
28
alternatives. Administrative Science Quarterly, 36:269-296.
29
30 Williamson, O.E. 1993. Calculativeness, trust, and economic organization. The Journal of Law
31 and Economics, 36(1, Part 2), 453-486.
32
33 Williamson, O.E. 1996. The Mechanisms of Governance. Oxford UK: Oxford University Press.
34
35
36
Williamson, O.E. 1999. Strategy Research: Governance and Competence Perspectives. Strategic
37 Management Journal, 20: 1087-1108
38
39 Yzerbyt, V. & Demoulin, S. 2010. Intergroup Relations. In Fiske, S.T., D.T. Gilbert, & G. Lindzey,
40 (Eds.), Handbook of Social Psychology, pp-1024-1083. Hoboken, NJ: John Wiley & Sons, Inc.
41
42
Zaheer, A., McEvily, B., & Perrone, V. 1998. Does trust matter? Exploring the effects of
43
44 interorganizational and interpersonal trust on performance. Organization Science, 9(2), 141-159.
45
46 Zaheer, A., & Venkatraman, N. 1994. Determinants of electronic integration in the insurance
47 industry: an empirical test. Management Science, 40(5), 549-566.
48
49 Zaheer, A., & Venkatraman, N. 1995. Relational governance as an interorganizational strategy:
50
51
An empirical test of the role of trust in economic exchange. Strategic Management Journal,
52 16(5), 373-392.
53
54 Zenger, T.R., Felin, T., & Bigelow, L. 2011. Theories of the firm–market boundary. Academy of
55 Management Annals, 5(1), 89-133.
56
57
58 80
59
60
Page 81 of 90 Academy of Management Annals

1
2
3
4
5
Zhao, H., Luo, Y., & Suh, T. 2004. Transaction cost determinants and ownership-based entry mode
6 choice: A meta-analytical review. Journal of International Business Studies, 35(6), 524-544.
7
8 Zhou, K.Z., & Poppo, L. 2010. Exchange hazards, relational reliability, and contracts in China:
9 The contingent role of legal enforceability. Journal of International Business Studies, 41(5),
10 861-881.
11
12
13
Zhu, F., & Liu, Q. 2018. Competing with complementors: An empirical look at Amazon. com.
14 Strategic Management Journal, 39(10), 2618-2642.
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 81
59
60
 Academy of Management Annals Page 82 of 90

1
2
3
4
5
FIGURE 1: WILLIAMSON’S TRANSACTION COST THEORY FRAMEWORK
6
7 BEHAVIORAL ASSUMPTIONS
8
9 Bounded Rationality
10 Opportunistic Behavior
11
12
13
14 ORGANIZATION
15
16
17
18
19
20
21 Transaction Characteristics Governance Mechanisms
22  Asset specificity  Markets
23
24  Uncertainty Alignment between the transaction  Hierarchies
25  Frequency characteristics and governance choice to  Hybrids
26 minimize transactions costs which arise due
27 to bounded rationality and the possibility of
28
29
opportunistic behavior
30
31
32
33 PERFORMANCE
34
35
36
A function of appropriate alignment of governance mechanism with transaction characteristics
37
38
39
40
41
42
43 82
44
45
46
47
Page 83 of 90 Academy of Management Annals

1
2
3 FIGURE 2: CURRENT/FUTURE APPROACHES TO TCT
4
5
6 BEHAVIORAL ASSUMPTIONS
7 Bounded Rationality
8 Opportunistic Behavior
9
Bounded Reliability
10
11
ORGANIZATION
12
13
14 Transaction Characteristics /
15 Characteristics of Interdependence Governance Mechanisms/Institutions
16  Market thickness Alternative Views
17 o Asset specificity  Markets
18 o Network effects  Hierarchies
19
20
o Other sources of thin markets  Hybrids
21 Alignment between the transactional o Contractual provisions – Control vs.
22  Behavioral uncertainty characteristics and governance choice to coordination
23 o Information asymmetry maximize transaction benefits which arise
24 o Appropriability due to bounded rationality and the
25 possibility of opportunistic or unreliable  Price system (output control)
26 o Output measurement costs
o Behavior measurement costs behavior  Hierarchy (behavior control)
27
28 o Pure markets
29  Environmental Uncertainty o Pure hierarchies
30 o Institutions that mix markets and hierarchies
31 (e.g., hybrids, relational governance)
32  Frequency
33
34
35
36 PERFORMANCE
37
38 A function of appropriate alignment of governance mechanism(s) with transaction characteristics given the micro (property rights, psychological framing) and
39 macro (polity, culture) contexts
40
41
42
43 83
44
45
46
47
 Academy of Management Annals Page 84 of 90

1
2
3 TABLE 1: THE BEHAVIORAL ASSUMPTIONS
4
5
6 Bounded Rationality
7 Issues Opportunities for Future Research
8 The bounded rationality assumption clearly There is substantial potential to integrate insights
9 acknowledges limits on the rationality of decision- from psychology and behavioral decision making
10 making, making TCT well-positioned to adopt a into TCT. Specifically, future research drawing on
11 more behavioral focus. Yet, relatively few scholars work on heuristics, attention, risk preferences, and
12 have embraced a more behavioral focus. biases (e.g., Kahneman et al., 1982; Ariely, 2008;
13 Gigerenzer & Gaissmaier, 2011; Kahneman, 2011)
14 would improve our understanding of how behavioral
15 uncertainty is perceived and incorporated into
16 decision-makers’ analyses.
17
18 In the literature on TCT, there are differing A greater emphasis on trying to understand why or
19 interpretations of the extent to which rationality is when some firms end up making governance
20 bounded, causing theoretical tension. On the one decisions that accord with the prescriptions of TCT,
21 hand, bounded rationality constrains actors’ ex ante and when firms make choices that are suboptimal
22 forethought sufficiently for transaction costs to from a TCT perspective, would address some of the
23 emerge. On the other hand, bounded rationality is not theoretical tension that is present in the TCT
24 sufficiently constraining to systematically hamper literature.
25 decision makers’ ability to make appropriate
26 governance choices in the face of future unknowns.
27
28
29
30 Opportunism
31 Issues Opportunities for Future Research
32 TCT has been criticized by some scholars for having Scholars in the international business literature have
33 a too pessimistic view of human nature. (As noted in proposed less extreme versions of opportunism:
34 the text, this is sometimes driven by an overly “bounded reliability” (Verbeke & Greidanus, 2012).
35 pessimistic interpretation of the definition of We see value in considering different views of
36 opportunism.) human nature and exploring whether these can be
37 reconciled with the logic of TCT and relevant
38 findings.
39
40 TCT has not fully exploited opportunities to integrate Insights from behavioral economics and from
41 behavioral insights to improve our understanding of different strands of psychology (e.g., Roth &
42 the opportunism assumption. Murnighan, 1982; Kahneman, Knetsch, & Thaler,
43 1990; Loewenstein et al., 1993) can help future TCT
44 research to better illuminate how and when decision-
45 makers perceive the potential for opportunistic
46 behavior and how this drives their governance
47 choices.
48
49
50
51
52
53
54
55
56
57
58 84
59
60
Page 85 of 90 Academy of Management Annals

1
2
3 TABLE 2: TRANSACTION CHARACTERISTICS AND GOVERNANCE MODES AND
4
5
MECHANISMS
6
7 Transaction Characteristics – Asset Specificity
8 Issues Opportunities for Future Research
9 Many TCT studies present a generic argument Future work should place more emphasis on identifying
10 pertaining to asset specificity, without considering the what type of asset specificity is relevant in the specific
11 details of their setting and the characteristics of the context that is studied and be more precise about how
12 investment that is made. that particular type of asset specificity drives
13 governance choices.
14
15 Studies using secondary data to capture asset specificity Future empirical research on asset specificity would
16 have often used measures that are potentially benefit considerably from using finer-grained data (i.e.
17 problematic. Namely, R&D and marketing intensity at the firm level or the transaction level itself) and more
18 data, often measured at the industry-level, have been precise and direct proxies for asset specificity (e.g.,
19 commonly used to operationalize asset specificity, Stuckey, 1983).
20 which might lead to inaccurate findings.
21
22 Studies using primary/survey data often fail to directly Researchers collecting primary data should design
23 capture the essence of the asset specificity construct, surveys with the precise theoretical construct more
24 since they assess the amount of investment that is made firmly in mind, and also collect data on the value of the
25 but not the extent to which these investments would investment outside the current use, i.e., its residual-
26 have residual value outside of the current transaction. value, or even measure the amount of quasi-rents at
27 risk.
28
29 The vast majority of TCT studies has focused on the Where possible, future studies should aim to collect and
30 asset-specific investments made by one exchange party use data on all the asset-specific investments of all
31 only, ignoring the asset-specific investments made by exchange parties. This would reduce the likelihood of
32 the other party. Empirically this may generate an biased findings, and would also allow the study of
33 omitted variable bias and theoretically it might additional theoretical issues such as “credible
constrain researchers from addressing more dyadic commitments” and the dynamics of the commitments
34
questions. made by all parties in the exchange.
35
36
Digital transformation and other economic trends have Scholars should assess more systemically whether new
37
a profound impact on the types of investments and types of investments fall within the existing categories
38
commitments firms make. TCT has insufficiently of asset specificity, or whether such investments
39
explored how this might affect existing constitute a new category altogether and require
40 conceptualizations of the asset specificity construct. conceptual refinements of the asset specificity
41 construct.
42
43
44 Transaction Characteristics – Environmental Uncertainty, Behavioral
45 Uncertainty and Appropriability
46 Issues Opportunities for Future Research
47 The international business literature has proposed Future research could theorize more precisely about
48 various sources of behavioral uncertainty that might and improve our understanding of the behavioral
49 matter for governance choices. However, there has been uncertainty concept by categorizing, in a theoretically
50 a lack of systematic categorization of these factors. coherent manner, the various mechanisms that have
51 been linked to behavioral uncertainty.
52
53 In empirical work there is a lack of clarity about when In deciding whether to use subjective or objective
54 and whether researchers should use objective or measures, researchers should carefully consider the
55 subjective measures of behavioral uncertainty. objective of the particular study: When the objective of
56 a study is to establish how firms should make decisions
57
58 85
59
60
 Academy of Management Annals Page 86 of 90

1
2
3 from a TCT perspective objective measures might be
4 more appropriate, when the objective of a study is to
5 explain deviations from TCT predictions, or to improve
6 our understanding of how decision-makers actually end
7 up making decisions that are based on TCT logic,
8 perceptual measures might be more useful.
9
10
11 Transaction Characteristics – Frequency
12
13 Issues Opportunities for Future Research
14 Frequency has received little attention, even though it More work is needed that systematically considers the
15 is an inherent part of classic TCT. Empirical effect of frequency.
16 operationalization has varied between continuous and
binary (one-time vs. recurrent transactions) measures of
17
frequency, with different patterns of results.
18
It is difficult to disentangle Williamson’s (1985) Researchers should try to use more refined empirical
19
proposed cost-amortization mechanism, which would methods to support better identification of the effects of
20
favor hierarchical governance of recurrent transactions, frequency. For example, they should aim to (i) directly
21 from alternative mechanisms (e.g., trust, the capture the cost implications of more-frequent
22 development of routines) that would favor non- transactions, which are typically measured as a latent
23 hierarchical governance of transactions. construct, (ii) control more systematically for other
24 mechanisms that are likely at play, (iii) use more
25 controlled settings where frequency can be manipulated
26 to allow for a more precise examination of its effect and
27 to rule out alternative explanations.
28
29
30
31
Governance Modes and Mechanisms: Markets, Hierarchies and Hybrids
32 Issues Opportunities for Future Research
33 A number of alternative views on governance have There is a need for more careful theorizing and
34 emerged that have implications for “hybrids” and the empirical investigation of the governance properties of
35 market-hierarchy continuum. different governance alternatives.
36
37
38 Governance Modes and Mechanisms: Relational Governance
39 Issues Opportunities for Future Research
40 There are diverging views on whether relational Future research should try to achieve further and more
41 governance is a third type of governance mechanism coherent theoretical integration of the mechanisms
42 alongside markets and hierarchies, or whether it refers proposed in the relational governance literature within
43 to a set of social mechanisms that influence behavioral the TCT framework.
44 uncertainty within the TCT framework.
45
46 Many studies in the TCT literature focus on the Anglo- More work is needed studying governance in parts of
47 Saxon world, where contracts and courts play a pivotal the world where transactions are less commonly
48 role in governing transactions. governed by contracts or courts.
49
50
51
52
53
54
55
56
57
58 86
59
60
Page 87 of 90 Academy of Management Annals

1
2
3 TABLE 3: TCT LOGIC
4
5
6 Issues Opportunities for Future Research
7 Different versions of TCT have evolved: Work in There is an opportunity to integrate and reconcile the
8 strategy has focused largely on market failure and logics underlying these extensions and alternative
9 appropriability while work in international business views to develop a more pluralistic version of TCT.
10 has put more emphasis on behavioral uncertainty and
11 why firms might fail.
12
13 Theoretical and empirical research has not fully Future research can address a number of questions
14 exploited opportunities to employ a dynamic that are particularly relevant in today’s competitive
15 approach to TCT. landscape by taking a more dynamic approach. For
16 example, future research could:
17  Investigate the role played by managers’
18 foresight and their understanding of the
19 “fundamental transformation” in their
20 governance choices.
21  Explore the adaptation of misaligned
22 governance modes over time, specifically
23 looking at finer-grained governance
24 adaptations, and at the effect of internal
25 factors that yield such changes.
26  Study the actual adaptation process to
27 illuminate and how firms adapt their
28 governance modes.
29
30 The TCT literature has predominantly focused on Scholars can explore the performance implications
31 explaining strategic choices, with less effort devoted of governance choices using a two-step approach:
32 to the performance implications of these choices.  Step 1 consists of careful theorizing about
33 what the appropriate choice for a firm is, and
34
then incorporating this theorizing and
35
identification in a research design, so that
36
one can reliably predict what a firm’s
37
governance choice should be in the first
38
place. This allows researchers to assess how
39
well-aligned or misaligned the firm’s actual
40
choice is.
41
 In step 2 the identified level of
42
(mis)alignment can be used to predict
43
performance differences.
44
Future research could also investigate the conditions
45
under which the performance implications of
46
misaligned governance choices for firms are more or
47
less severe.
48
49
50
51
52
53
54
55
56
57
58 87
59
60
 Academy of Management Annals Page 88 of 90

1
2
3 TABLE 4: THE FUTURE OF TCT
4
5
6 PANEL A: CONCEPTUAL EXTENSIONS
7
8 1. Further integration of strategy, international business, and institutional economics
9
10 Theory/Literature Opportunities for Future Research
11 More engagement with the international  How do cultural and institutional factors influence the
12 business literature choice between the enforcement of transactions through
13 relational governance vs. through contracts?
14  How do cultural and institutional factors influence
15 contract design?
16
 How does institutional variation across countries
17
18
influence the enforceability of contracts?
19  How do cultural and institutional differences across
20 countries affect the reliance on contracts vs. relational
21 governance mechanisms?
22  How do linguistic factors influence the design and
23 effectiveness of contracts?
24 More engagement with the strategy literature  How and when does prior experience influence
25 governance choices?
26  How do capabilities and adjustment costs influence the
27
incentives to adjust misaligned governance structures?
28
29  How do the control and coordination aspects of contracts,
30 or other governance mechanisms, interact?
31 Reconciling the strategy and international  How can apparent contradictions between alternative
32 business literatures views in strategy and international business be resolved?
33 Engagement with information economics and  How do different types and sources of information
34 signaling theory asymmetry drive behavioral uncertainty?
35
36
37 2. Link to sociology – trust/opportunism; formal/informal organization
38
39 Theory/Literature Opportunities for Future Research
40 Engagement with sociology:  To what extent, and where, is trust misplaced?
41  Trust and opportunism/behavioral  Can an understanding of opportunism help tease apart
42 reliability justified from unjustified trust?
43  Formal and informal organization  Can TCT illuminate mechanisms that can enhance trust
44 accuracy?
45
 To what extent is competence trust strategic?
46
 To what extent can insights from trust open up the black
47
48 box of TCT’s “shift parameters?”
49
50
3. Link to psychology/behavioral economics – context and framing
51
52 Theory/Literature Opportunities for Future Research
53 Engagement with psychology:  How does the framing of an exchange affect governance
54
 Framing and execution?
55
56
 Biases and Heuristics
57
58 88
59
60
Page 89 of 90 Academy of Management Annals

1
2
3  Attention  How does the framing of a governance structure affect
4
the way that it is perceived by actors?
5
6
 How does the process by which an agreement is reached
7 affect its subsequent execution?
8  How do heuristics, attention, risk preferences, and biases
9 affect the extent to which decision-makers evaluate the
10 potential for opportunistic behavior?
11 PANEL B: EXTENSIONS TO NEW PHENOMENA
12
13 1. Technological advance – platforms and governance
14
15 Phenomenon/topic Opportunities for Future Research
16 Platforms, ecosystems, two-sided markets  How does a platform’s ability to exclude parties affect
17 the governance of exchanges that occur?
18  Does a technology platform represent an alternative
19
form of governance?
20
21  How does the threat of market tipping affect market
22 thickness and governance choices?
23
24
25 2. Technological advance – artificial intelligence, machine learning, and transaction costs
26 Phenomenon/topic Opportunities for Future Research
27
Artificial intelligence/machine learning; the  How do technological advances affect the costs of
28
29
economics of prediction; technology and work monitoring/enforcement/ coordination?
30
31 3. Non-pecuniary phenomena – nationalism; sustainability/corporate social responsibility;
32
33
government action via public-private partnerships; and “grand challenges”
34 Phenomenon/topic Opportunities for Future Research
35 Nationalism  How does nationalism affect opportunistic behavior and
36
(the perception of) behavioral uncertainty?
37
38 CSR  What are the most suitable types of monitoring for
39 different types of CSR activity?
40  What are the tradeoffs in encouraging CSR behavior?
41  Which organization form – for-profit-firms, not-for-
42 profit firms, NGOs, etc. – is more effective for which
43 types of social effort?
44 Public-Private Partnerships  When/how can government work effectively with
45 private-sector organizations?
46
 How does contract rigidity in the public sector affect the
47
48 utility of PPPs?
49  How do rigid civil-service rules affect the effectiveness
50 of “fiat” in the public sector?
51 Grand Challenges  How should collaborations to tackle grand challenges,
52 which are often more diverse and complex, be
53 governed?
54
55
56
57
58 89
59
60
 Academy of Management Annals Page 90 of 90

1
2
3 Ilya Cuypers ([email protected]) is an Associate Professor of Strategy at the Lee Kong
4
5
Chian School of Business, Singapore Management University. He received his Ph.D. in strategy
6 and international business from Tilburg University. His current research focuses on the governance,
7 dynamics and performance implications of cross-border external corporate development activities,
8 investment decisions under uncertainty, and issues related to interorganizational ties.
9
10
11
Jean-François Hennart ([email protected]) is Professor of International Management emeritus
12
13
at Tilburg University and visiting Professor at Politecnico di Milano and Aalborg University. His
14 research focuses on the comparative study of international economic institutions such as
15 multinational firms and their contractual alternatives, family firms, Born Globals, joint ventures
16 and alliances, and modes of foreign market entry and exit. His Theory of Multinational Enterprise
17 pioneered the application of transaction cost theory to international business. He is a Fellow of the
18 Academy of International Business and of the European International Business Academy and
19
holds an honorary doctorate from the University of Vaasa. His publications in the Journal of
20
21 International Business Studies, Entrepreneurship Theory and Practice, Strategic Management
22 Journal, Global Strategy Journal, Organization Science, Management Science, the Journal of
23 Economic Behavior and Organization and other major journals have garnered more than 18,000
24 Google citations.
25
26
27
28
Brian S. Silverman ([email protected]) is the J.R.S. Prichard Chair in Management,
29 and Professor of Strategic Management, at the University of Toronto’s Rotman School of
30 Management. His research focuses on the ways in which firms’ strategies and organizational
31 structures interact to affect their performance.
32
33
34
Gokhan Ertug ([email protected]) (PhD, INSEAD) is professor of strategic management
35
36
and Lee Kong Chain Fellow at the Lee Kong Chian School of Business, Singapore Management
37 University.
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 90
59
60