Thanks to prospect theory, economists have come to recognize that the assumption that individuals behave as rational actors maximizing utilities is not descriptively accurate. Do you believe that political scientists have adequately adjusted their own approaches given what we now know about human behavior?
Prospect theory runs counter to traditional economic rationality, in that according to an economic rational choice model, actors have fixed preferences that they are able to rank order. Therefore, given a decision situation, there is a clear order of what decision would give the best outcome, and there is no variance in this preference ranking.
Prospect theory, however, shows that the rank-order for preferences varies according to which decision frame an actor is operating in. When an actor frames a decision from a domain of loss, s/he is more likely to engage in risky behavior. Likewise, when an actor is in a domain of gain, s/he is less likely to engage in risky behavior. This concept was introduced and further developed by Kahneman and Tversky 1979/1984, and then Kahneman, Knetsch and Thaylor, who found evidence for a status quo bias, where individuals have a strong tendency to remain in the status quo, as well as the endowment effect, in which individuals demand more to give up an object than they would be willing to pay to acquire it in the first place. Another tenet of prospect theory is that people are loss averse, and more risk-acceptant to hold onto what they already have. Loss looms larger than gain. This is not rational, as economist understand rationality.
Prospect theory has received pride of place as a social scientific “law” if there is such a thing, and Kahneman received the Nobel Prize in Economics with this groundbreaking finding that, in fact, it is not descriptively accurate to assume that individuals behave as rational actors maximizing utilities. In economics, where price point is everything, and this departure from rational utility maximization is easily recognized (when someone is willing to pay $20 to have someone mow their lawn, but unwilling to mow someone else’s lawn for $25) the findings of loss-aversion and risk-acceptance have a direct and quantifiable bearing on theory. In international relations, however, the implications of prospect theory are less clear cut.
In all situations of international crisis and decision making, for example, there is significant risk involved. Often, the choices are between risky, riskier, and riskiest, not safe, status quo, or risky. Therefore, applying Prospect theory to IR is no less problematic than applying rational assumptions.
Additionally, in IR there are multiple facets of risk and variables to consider. It is not merely an economic unit of dollars of Euros to gain or lose and frame in a domain of loss and gain, but IR variables, in addition to military spending, include land, sovereignty, security, human lives, and reputation. All of these combine and come together in international decision considerations. There are multiple frames to consider; is the actor loss-averse in relation to land mass, good reputation, human lives, military capabilities, etc. It is not as simple as dollars and cents, which is the primary variable of economics. This is another reason why applying the implications of prospect theory proves difficult for IR scholars.
Another problem with applying prospect theory to international relations is that there is no way to tell a priori what domain the actor is operating out of: the domain of loss or gain (Mercer 2005). Based on the multiple ways that the variables could be framed (loss of life, land, ideals, etc) it is impossible to determine the frame of the decisionmaker. To know this requires a cognitive theory that accounts for affect, personality, beliefs. Once the frame is determined by a prior theory, then prospect theory can be applied. However, on its own, prospect theory is generally unuseful unless the decision situation is clear and simple, such as whether a state will respond to an attack on its land (to which it is risk-acceptant) or whether it will act aggressively to secure another state’s territory (to which is will be risk-averse). If however the situation has any added complexity or dimensions such as whether the land is of sacred value, or there is ideological components to the battle such as is the case in the many regions of the Middle East and South Asia, then a purely economic calculation of gains versus loss does not allow for a simple determination as to whether an actor is in a frame of loss or gain. Therefore, prospect theory on its own is not a complete model that can be adopted seamlessly into IR. Much prior theorizing is necessary, and the application of prospect theory at the end is then appropriate.
Another significant problem with applying prospect theory to IR is that it has the exact same problem that rational choice theory has. It has no theory of preferences. It is very possible, and very likely that everyone is subjectively, boundedly rational given their preferences. Prospect theory does show that preferences are not fixed and cannot be exogenously assumed. However, that does not help in the effort to determine the substance of the preferences. Perhaps an actors decision to concede one thousand acres to keep fifty acres is not “rational.” But the preference may not be what rationalists assume would be “retain the most land mass possible” but rather “preserve the sacred land with our ancestral burial grounds or a holy site.” Another example involves path dependency and sunk costs. Perhaps the state preference is to continue a losing war for the moral fulfillment it brings, or continues on the path because pulling out would imply that the lost troops “died for nothing.” Therefore, one cannot disprove actor rational utility maximization until one knows how preferences are formed and know the substantive content of the preferences. Without this critical knowledge, neither prospect or rational choice theories can perform. Without a clear knowledge of preferences in the first place, it is not possible to say that prospect theory nullifies rational utility maximization. Therefore, a “law” of Economics does not translate to political science or any field where preferences are problematized. In the social sciences, what may appear to be “irrational” is really the rational behavior of an actor seeking to maximize a goal that the scientist misspecifies.
Political science has not adequately answered the question that is of ultimate value here: where do preferences come from, and how can we know an actors preferences a priori? The studies in IR that have used prospect theory assumed both the actors’ frame and preferences after the fact, which is problematic. For example, Farnham claims that moving from a domain of gain to a domain of loss caused FDR to become more risk-acceptant and led the US into the fray of World War II. She does a good job of controlling for the admission of new information that would change the expected utility, but still, her claim that FDR’s framing of US intervention in the war changed and led him to become more risk-acceptant is based on the post hoc fact that the US did indeed intervene in the escalating conflict. There is a problematic lack of looking at preferences as a dependent variable.
Ultimately, a lot of actions that are deemed “irrational” in IR are not necessarily irrational at all. For example, the suboptimal (sub-rational) outcomes associated with groupthink would not be considered irrational at all if the preference to be maximized was understood to be group cohesion and unanimity (Janis). Additionally, if ideational and normative concerns are included in the substantive considerations of a “preference” many actor decisions may be seen not as non-rational, but as fitting perfectly within a rational framework. The real failure in rational and prospect theory is the same: a lack of theorizing preferences, and in prospect theory, the lack of theorizing about the frame.
No comments:
Post a Comment