Navigation page

Pages

Saturday, September 10, 2011

More on meso causation

A recent post considered the question, do organizations have causal powers? There I argued that they do, in a number of ways. Here I'd like to return to these claims and see how they disaggregate onto subvening circumstances, including especially patterns of individual and group activity. The italicized phrases are extracted from the earlier post.
  • First, the rules and procedures of the organization may themselves have behavioral consequences that lead consistently to a certain kind of outcome.
How do rules and procedures causally affect the behavior of the actors who participate in them? (a) Through training and inculcation. The new participant is exposed to training processes designed to lead him/her to internalize the procedures and norms governing his/her function. (b) Through formal enforcement. Supervisors are institutionally charged to enforce the rules through direct observation and feedback. (c) Through the normative example of other participants, including informal sanctions by non-supervisors for "wrong" behavior. (d) Through positive incentives administered by supervisors and mid-level functionaries. Each of these avenues for influencing the behavior of an actor within an organization depends on the actions and motivations of other actors within the organization. So we have the recursive question, what factors influence the behavior of those actors? And the answer seems to be: all actors find themselves within a dynamic system of behavior by other actors, frequently maintaining an equilibrium of reproduction of the rules and roles.
  • Second, different organizational forms may be more or less efficient at performing their tasks, leading to consequences for the people and higher-level organizations that are depending on them.
Institutions designed to do similar work may differ in their functioning because of specific differences in the implementation of roles and processes within the organization. This is a system characteristic of the particular features and interactions of the rules and processes of the organization, along with the expected behaviors of the participants. It is also a causal characteristic: implementing system A results in greater efficiency at X than implementing B. The underlying causal reality that needs explanation is how it comes to pass that participants carry out their roles as prescribed--which takes us back to the first thesis.
  • Third, the discrepancy between what the rules require of participants and what the participants actually do may have consequences for the outputs of the organization.
This causal claim highlights the difference between formal and informal procedures and practices within an organization. Informal practices can be highly regular and reproducible. In order to incorporate their implications into our analysis of the workings of the organization we need to accurately understand them; so we need to do some organizational ethnography to identify the practices of the organization. But in principle, the logic of explanation we provide on the basis of informal practices is exactly the same as those offered on the basis of the formal rules of the organization.
  • Fourth, the specific ways in which incentives, sanctions, and supervision are implemented differentiate across organizations.
This is one of the key insights of the "new institutionalism." The specific design of the institution in terms of opportunities and incentives presented to participants makes a large difference in actors' behavior, and consequently a large difference to the system-level performance of the institution. Tweaking the variable of the level in the organization's hierarchy that needs to sign off on expenditures at a given level has significant effects on behavior and system properties. On the one hand, higher-level sign-off may serve to restrain spending. On the other hand, it may make the organization more unwieldy in responding to opportunities and threats.
  • Fifth, the organization has causal powers with respect to the behavior of the individuals involved in the organization.
This factor parallels thesis 1 but is meant to refer to longterm effects on behavior and personality. The idea here is that immersion in a particular organization and its culture creates a distinctive social psychology in the people who experience it. They may acquire habits of thought, ways of responding to new circumstances, higher or lower levels of trust of others, and so forth, in ways that influence their behavior in the broader society. The idea of an "organization man" falls in this category of influence. The organization influences the individual's behavior, not just through the immediate system of rewards and punishments, but through its ability to shape his/her more permanent social psychology.

There are only two fundamental causal paths identified here. The causal properties of the organization are embodied in the patterns of coordinated actions undertaken by the actors who are involved; and these orderly patterns create system effects for the organization as a whole that can be analyzed in abstraction from the individuals whose actions constitute the micro-level of the social entity.

The most obvious causal property of an organization is bound up in the function of the organization. An organization is developed in order to bring about certain social effects: reduce pollution or crime, distribute goods throughout a population, provide services to individuals, seize and hold territory, disseminate information. These effects occur as a result of the coordinated activities of people within the organization. When organizations work correctly they bring about one set of effects; when they break down they bring about another set of effects. Here we can think about organizations in analogy with technology components like amplifiers, thermostats, stabilizers, or surge protectors. This analogy suggests we think about the causal powers of an organization at two levels: what they do (their meso-level effects) and how they do it (their micro-level sub-mechanisms).

No comments:

Post a Comment