Analytical Paper 1 This week’s reading and discussion were guided by several

Analytical Paper 1

This week’s reading and discussion were guided by several underlying questions about how individuals and groups behave. Namely, what is, if there is, the logic behind decision-making? Is the decision-making calculus different or not between groups and individuals? If so, why? And, what does our method of understanding behavior tell us about fundamental questions of international political economy? From this, the readings and discussion tried to present a coherent foundational building block to understanding these questions; however, such a staunch paradigmatic focus on a singular explanation leaves out too much and does not appropriately address the weaknesses of the approach presented. To get at this, this essay is broken down into two separate components. The first component will cover, briefly, what was presented during the course of the reading and the following discussion. The second component will go more in-depth on the argument I have brought up.

What We Read and What We Know So Far

The formative question that allows for further discussion is how are decisions made? What the readings, mostly, and the discussion, entirely, focused on was a rational-choice perspective of decision-making. That is, individuals adhere to some coherent rationale based on pursuing actions that seek to maximize their rank-ordered and transitive preferences. Such that when presented with multiple choices, an individual will have a consistent behavior when making choices relying on the order of their transitive preferences. This behavioral position is the starting point for Coase’s (1972) argument for an increase of government regulation in the ‘market of ideas’, Akerlof’s (1979) argument for how uncertainty influences consumer choice, and Olsen’s (1981) logic of collective action. What can be seen is that the rational-choice model is a pervasive and powerful model of understanding decision-making.

From an individual perspective, rational choice modeling makes intuitive sense. Each individual will have to be presented with a myriad of choices in their day ranging from the mode of transportation to their work to who they will vote for come election season. The underlying assumption is that there is some coherent logic that ties an individual’s decisions together based on the choices and the incentives for those choices presented to them. However, this begs the question of whether these individual choices, when aggregated, are reflected in group decision-making.

The answer, according to Olsen (1981), is very clearly no. Individuals are unique with their own list of preferences which creates instances, that even if group decision-making can occur (i.e., free-riding has been overcome), where groups pursue sub-optimal choices and act in ways that are, at face value, irrational. Before introducing potential incoherences in the parts of the logic of how Olsen gets here, the value of this approach demonstrates that groups function separately from the individuals that compose them.

To better understand this we can explore two examples. The first is derived from the famous Condorcet voting paradox wherein a minimum of three individuals are chosen to elect an individual by unanimous decision, but each with their own unique order and transitive preferences will come to no decision at all. Because of the transitive preference orders of the three individuals, there will always be one individual who will disapprove of the potential decision thus reducing the ability for the group to make a decision. While each individual, on their own, would be able to come to a decision with a coherent logic, as the number of individuals increases the likelihood of a coherent group decision being made decreases. Condorcet here is essentially leading the way for Olsen’s (1981) logic to take hold.

The second example comes from Olsen’s fourth implication of group decision-making. The implication is that special interest groups are more likely to reduce the overall income level of the countries they exist in to attain a greater share of the present income distribution. What is irrational here is that individuals, on their own, would be seeking to increase not only their share of the proverbial “pie” but also their total overall well-being. However, under Olsen’s analysis, these special interest groups are more incentivized to reduce overall income to increase that group’s share of the total income even if it harms individual members of that group. Because the group decision-making calculus is different from individuals based on not only the inability to form coherent decisions (see Condorcet voting paradox) but also that groups have different incentives, at times, than individuals do.

Finally how this understanding influences our understanding of international political economy comes in two forms. The first is in understanding the exogenous and endogenous influences of these choices. One portion of the discussion centered around the formal and informal institutions that shape individual and group decisions. A type of informal institution that was discussed was class. The argument goes that class and incentives that are produced by it influence the ability to make collective decisions. Beyond Marxist conceptions of the bourgeois and the proletariat, class is stratified by both income level and type of labor individuals participate in. That a middle-class office worker has different incentives and group belonging than a middle-class worker in the trades. While both belong to the same economic class, they have different localized incentives that might compete with broader class interests. Essentially, beyond the formal institutions of the state or unionized labor, informal institutions will produce further competing incentives thus influencing the rational decision-making calculus of individuals and groups.

The second comes from a quick discussion on the global “tragedy of the commons” which is climate change. The argument originally put forth was that the garment trade between Italy and Bangladesh was seen as a failure of collective action on climate change due to the high amounts of pollution and greenhouse gases produced specifically in this industry. However, depending on how one conceptualizes a group the incentives for that group will vary. For example, if the group is defined by the states involved in and dependent on the garment industry, it is entirely logical to support trade between those that produce and distribute those goods. But if the group was defined as the global community, then further exacerbating climate change through propping up the garment industry could be seen as a collective action failure. How one defines the group influences the ability to understand coherent versus incoherent group decisions.

More than this, understanding the internal preference structures of groups will allow us to understand how and why these groups do or do not come to decisions. Beyond free-riding, certain actors are incentivized to actively go against what the majority portion of the group prefers. Returning to climate change, the argument that not aiding in stopping climate change is a collective action failure assumes that all states are equally threatened by climate change. This is not the case, there is a spectrum of climate winners (those who benefit from climate change – ex. Russia) to climate losers (those who are harmed by climate change – ex. Bangladesh). It becomes more apparent to understand group decision-making is dependent on understanding blocs of competing incentives within groups. Understanding the totality of individual preferences is almost, if not entirely, impossible, but there is a potential to understand the preferences of sub-groups within groups to better understand group decision-making. Essentially, if we can know the competing interests within groups we can better understand the group’s overall decision.

Where Are We Lacking?

So far, based on a rational choice perspective, we are able to create a powerful and coherent understanding of individual and group behavior. From a standpoint of rank-ordered and transitive preferences we are able to explain, at least in part, how behavior is either rational or irrational. However, the entire discussion not only sidestepped but ignored the powerful point (and a whole topic of their work) raised by Gilpin and Gilpin (2001) that “Interpretations of economic affairs are highly dependent upon the analytic perspective of the observer and upon his or her assumptions as these determine what the observer looks for or emphasizes” (pg. 31). While we have a produced a solid pure neo-classical explanation of behavior we have inadvertently ignored non-neo-classical explanations. Gilpin and Gilpin (2001) use the example of the 1992 NBER report where the findings of the political scientists and economists differed not only due to the data used but also their underlying assumptions of the behavior of Japan. Since drastic differences in understanding can arise based on the assumptions placed into the model, the driving force of this portion of my essay is: using only a rational choice model to understand individual and group, what are we leaving out?

To get at this, I believe we can rely on Gilpin and Gilpin’s (2001) example of the 1992 NBER report to explore this difference. Gilpin and Gilpin state, simply, that because one focused on rational choice (economists) and the other on culture and nationality (political scientists) with the underlying assumptions of each led to differing conclusions. However, there is no discussion in Gilpin and Gilpin about why this is the case other than stating, with some conjecture, that the divergence in behavioral assumptions is the cause of different findings. I argue, that structural arguments (i.e., culture and nationality) when coupled with rational choice’s agent-centric approach, lead to a better understanding of individual and group behavior. This does not mean that one approach is better than the other, each with its merits and faults, but that a discussion of one without the other leads to a less-than-full understanding of individual, group, and state behavior in international political economy.

What about culture and nationality can bring an enhanced view of individual and group behavior? Compared to rational choice and its agent-centered focus, the focus of culture and nationality relies on understanding structure. The most minimal view of how individuals and groups operate is based on macro processes producing incentives. Such that based on material constraints and incentives, individuals and groups will act in accordance with those material structures. Take, for instance, individuals of a certain class, due to material interests, who will combine into larger structures to act on the structure-induced incentives.

Ironically, Olsen’s work is a direct critique of this viewpoint. Viewing material interests from the agent’s perspective (micro) rather than the structures perspective (macro), leads him to explore how there is, or is not, collection action. Individuals rarely act in coordination with others to gain the goods required by collective action. From a purely micro perspective and solid critique of the macro, there is little incentive for individuals to act together for mutual gain and a broad understanding of the seemingly irrationality of groups. However, there seems to be a contradiction between reality and Olsen’s propositions and within Olsen’s own work. These contradictions can only be explained through a more structural perspective.

To uncover these contradictions, I will be working with the latter contradiction first. In Olsen’s (1981) work, he posits simultaneously that individuals are unlikely to join together for collective action but the stability of the macro environment produces an ecosystem that makes collective action easier. The incentives of individuals, regardless of the environment according to Olsen, lead them to prefer free-riding over engaging in collective action thus making collective action a rare phenomenon. However, the stability of an environment produces more collective action. The argument for this is that stability reduces external costs to the individual and thus increases the relative cost-to-benefit ratio to the individual. However, there is no discussion, if not handwaving, to the fact of how this stability arises in the first place. The agent-centered focus cannot explain this without first assuming that collective action occurred in such a way that produced a positive stable social environment. However, implications 4-9 of Olsen would produce a social environment the opposite of stable with groups incentivized to reduce overall income, lack of technological innovation, slow economic growth, a divided society, and make future coordination difficult due to complex regulations that favor existing groups over emerging groups. In Olsen’s own logic, groups emerge during stable times, but stability is not a product of group behavior. Begging the question of how do these groups emerge? And, if there is a flaw in the fundamental notion of how groups emerge, what else is flawed? Using an agent-centric model that ignores larger macro-structures produces a skewed and contradictory theory of group emergence and behavior.

With understanding a few contradictions within Olsen’s work, I can now discuss how there is a divergence between theory and reality. To “Steel Man” the position, theory is, almost by definition, an abstraction of reality. There is always an incongruence between the model we produce and the world we are seeking to explain. But how much incongruence is allowed before the model is considered too false to be true? The answer lies not in whether the specifics of the model represent reality (a 100% pure predictive model) but if the assumptions of that model reflect reality. Models are generalities and thus the answers derived from those models should be general with the knowledge that outliers will occur. Models, however, only become generalizable if the assumptions are, in some way, reflections of the reality they are trying to capture.

The assumption in question from Olsen’s work is that because individuals are unlikely to, individually, move society much they are unlikely to acquire public information – be rationally ignorant (see his argument on pg. 26-27). Specifically, Olsen argues that because individuals, in a Downsian world, view voting as worthless or irrational, individuals are thus also disincentivized to seek out information on the election. But the underlying assumption of his observation (Downsian voting) is wrong. Riker and Ordeshook (1968) argue and find in their famous article “A Theory of the Calculus of Voting” that individuals have sociotropic concerns beyond themselves that produce the necessary incentives to vote. Voting, now being rational, would incentivize individuals to at least be minimally informed to make a decision that reflects their preferences and the preferences of their group. Thus making the internal decision-making structure of groups much more coherent. The downstream influences of this change in Olson’s assumptions go beyond what I can write here. However, the primary implication is that group behavior is much more rational and potentially reflective of individual preferences than what Olson would make it seem.

However, I should not rest on my laurels here. Taking Coase’s (1972) argument seriously, individuals are not incentivized in the truth but in the conflict between truth and lies thus presenting a market failure in the informational sphere. More than this, producers of information are also incentivized to sensationalize that information at the expense of honesty. Thus even if individuals are gaining information, it may be bad information that could produce circumstances similar to individuals who are rationally ignorant. But even with this slight reframing, this does not address the broader claim of the new assumption I posited. The assumption is that individuals will seek out information not only based on their own preferences, but also on the preferences of their group or broader context. Such that Democrats would seek out information, at minimum, that supports their a priori positions but the perceived positions of the larger party just as Republicans would the same. The reframing using Coase (1972) only makes Olson’s second statement more palatable, but it still relies on the Downsian perspective that information-seeking becomes irrational. Voting, in Riker and Ordeshook’s (1968) world, is rational because of the structural/sociotropic concerns of the individual. Replacing the underlying assumption with an update that better reflects reality (individuals have broader concerns and seek out group-insulating information) will produce inferences that will almost certainly better reflect reality as well. That is, ignoring the broader context (read structure) that individuals exist in and solely focusing on the individual themselves produces a skewed explanation based on faulty assumptions.

Thus far I have danced around what it is exactly about explicitly integrating structural approaches into the neo-classical rational choice model brings to our understanding of individual and group behavior. I have provided critiques of purely agent-centric approaches through more structural arguments, but still have not stated exactly why this integration improves upon what was discussed in the seminar and could be read from the readings. To be explicit, individuals exist in a broader social context that will induce and produce incentives for individuals that are subsumed into and beyond their individual preferences. This does not mean that these top-down pressures are the only influences, agents still exist and operate within their social ecosystem. Focusing completely on one side produces inferences that are skewed because they are based on faulty assumptions. From a purely structural perspective, Olson was able to powerfully critique structural arguments using an agent-centric model to produce models of collective action and group behavior. However, a purely agent-based model ignores key structural elements which produces a skewed theory of individual and group behavior.

While neither approach is without its fault, each brings important aspects to our understanding of behavior. Portions of the conversation held in class touched on this in the smallest way ( the discussions on emotions and institutions); however, the notion of synthesization between structure and agency is an important piece to better understand how behavior is affected. While rational choice is a powerful, cogent, and generalizable model of behavior, it cannot fully explain behavior. More than this, ignoring the broader context leads to faulty assumptions leading to even faultier inferences from those assumptions. To ensure that our models are able to explain, at least in part, reality is to have reality be reflected, in part, in our models.

Where We Are Left

This essay sought to do two things. The first is to establish an understanding of how the rational choice model views individual, group, and state behavior. Secondly, argue that a sole focus on a single paradigm that does not synthesize the merits of others produces incorrect conclusions about the very phenomena we are seeking to explain. The argument presented was not designed to denigrate a particular viewpoint, but rather show that tunnel vision reduces our overall explanatory power of behavior. The rational choice model is seen as a highly valuable tool to explain behavior for a reason – because it is. But it is not the be-all-end-all. To get at a better explanation requires either A) assessing the appropriate scope conditions that one’s approach can explain (not over-generalizing the argument and model) or B) synthesizing approaches to better reflect reality to produce more accurate inferences.

References

Akerlof, G. A. 1970. The Market for “Lemons”: Quality Uncertainty and the Market Mechanism. The Quarterly Journal of Economics, 84(3), 488–500.

Coase, R.H., 1974. The market for goods and the market for ideas. The American Economic Review, 64(2), pp.384-391.

Gilpin, Robert, and Jean M. Gilpin. 2001. “The Nature of Political Economy.” In Global Political Economy: Understanding the International Economic Order, Princeton University Press: pp. 25–45.

Mancur Olson, 1982. The Rise and Decline of Nations. Yale University Press:

Riker, William H., and Peter C. Ordeshook. 1968. “A Theory of the Calculus of Voting.” American Political Science Review 62(1): 25-42.