Beyond generics: A closer look at hybrid and hierarchical governance

This study is about disaggregating the generic modes of governance as they

are defined in Transaction Cost Economics (TCE). More specifically, it intends

to increase the level of resolution of TCE in the field of hybrid and hierarchical

governance by specifying (some of) the subcategories of governance within

these two generic modes and relating these subcategories explicitly to the

transactions they control. It is argued that such a disaggregation is useful for

two main reasons: (1) it may increase the accuracy of TCE's predictions and

may improve the expressiveness of its style of explanation, and (2) it may

enlarge the conceptual scope of TCE, opening up problem areas that previously

did not fit neatly into the realm of this approach.

This general idea ties together the two substantive parts of this study. The

first of these starts from the empirical observation that hybrid structures

sometimes survive conditions of substantial uncertainty -an observation that

does not go particularly well with received TCE-, and examines two cases of

hybrid contracting in such conditions. It is argued that both cases are examples

of a hitherto ignored subcategory of governance that, once identified,

restores TCE's ability to explain this observation. The second part brings

TCE's explanatory apparatus to bear on issues of management control. It is

shown that TCE supports a detailed study of control issues, and that it has

much to offer to the explanation of control structure variety within (and

beyond) the hierarchy.

Hybrid contracting and uncertainty

There is a growing body of empirical evidence showing that sometimes, hybrid

structures are chosen for transactions that combine substantial asset specificity

with significant uncertainty. This evidence meets uneasily with TCE.

Extant TCE suggests that for such transactions, hierarchical governance with

its distinctive blend of cooperation-inducing features and sequential adaptation

is uniquely suited. The hybrid form, on the other hand, is considered

viii

infeasible in conditions of uncertainty, because it requires a fairly complete

ex ante explication of the particulars of the transaction.

In this study, I argue that TCE's somewhat overstated position on the infeasibility

of hybrid contracting in conditions of uncertainty is best be rectified by

taking a closer look at the mechanisms of governance on which apparently

uncertainty-resistant hybrids rely. It may very well be that extant TCE puts

too much emphasis on compliance arrangements, and that there are in fact

different (configurations of) control mechanisms available to the hybrid form

to mitigate opportunism.

An analysis of two generalized cases of hybrid contracting in conditions of

asset specificity and uncertainty (outsourcing in the Japanese automobile

industry and venture capital financing) revealed the contours of a subcategory

of the hybrid mode that, unlike its more familiar compliance-focused counterpart,

allows substantial contractual incompleteness. This subcategory

invokes both market-based incentives and intensive exchange of information.

The market-based incentives foster behaviour congruence without requiring

performance goals or standards to be specified in advance, whereas information

exchange and the resulting transparency allow significant direct control

over the actions of the contracting partner during the process of contract

execution. The combination of the two facilitates harmonious interim adjustment

and correction, and in both cases, this configuration of governance

devices seemed an efficient solution to the relevant contractual problems.

Transaction Cost Economics and Management Control

One of the quintessential problems of management control (MC) as a field of

scholarly inquiry is to explain control structure variety within and between

organizations. However, previous theorizing in MC has not been able to address

this issue fully satisfactorily. In this study, I suggest that substantial

progress can be made by applying TCE to the issue at hand. MC shares its

central problem - explaining control- with TCE, albeit that the former requires

a higher level of resolution. The logic of TCE, however, is receptive to refinement,

and supports a detailed study of control issues at the level of organizational

subsystems. At that analytical level, I propose a transaction cost

ix

theory of MC. This theory specifies the composition of various archetypal

control structures, and links these to the kind of activities they are expected

to control.

The argument runs as follows. The nature of the organizational activities and

the contributions from organizational participants that are required to perform

these activities can be defined discriminatingly through their scores on

three dimensions: (1) the extent to which the contributions are susceptible to

up front programming; (2) the degree of asset specificity; and (3) the intensity

of ex post information impactedness. Given bounded rationality and

opportunism, these features are predictably associated with distinctive control

problems that need to be dealt with. The various control archetypes

differ in their problem-solving ability, which makes them appropriate for the

governance of some contributions, but not for others. Moreover, they differ in

respect of cost, and ultimately, an empirically observable alignment of a

contribution with a control archetype can be explained by delineating the

relative efficiency properties of the match, either quantitatively or - more

likely- in a qualitative way.

This theoretical approach has some qualities that make it worth considering.

For one, it is empirically testable. Furthermore, its relatively simple theme

seems to speak to a wide empirical domain, and can be used to make sense of

a large set of remarkably different control structures in a consistent and

coherent way. And finally, the proposed theory offers a practicable procedure

to handle the issue of defining the organizational goals that MC is supposed to

serve, and an operational way to address control structure effectiveness.