Hey everyone. Noah Zerbe here. This is one
of a series of videos looking at decision making in foreign policy. In this video, we're
going to break down the rational actor model. So let's start with the question of what makes
an actor rational? That is, what do we mean by rationality? Graham Allison defines rationality
as, "consistent, value maximizing choices, with specified constraints." If we understand
rationality in this way, we see that rationality is actually built on a few assumptions.
First, we assume that action is purposive. That is, actors engage in specific
actions or make specific decisions in pursuit of a particular goal. Decisions
are not made by habit or social expectation, but instead pursue a specific objective. Second,
we assume that the goal itself is clearly defined before the decision is made. That is, goals exist
a priori to decisions. The agent must be able to identify their ultimate objective in order to act
to achieve that objective. Third, we assume that preferences or goals are ranked and exist with
a transitive order. That is, we assume that from any individual decision maker's perspective,
they've rank ordered their preferences: their first choice, their second choice, their third
choice, and so on, and that if a decision maker prefers outcome one to outcome two and outcome
two to outcome three, then they would also prefer outcome one to outcome three. Rationality also
assumes that preferences are invariant. That is, we assume that our preferences hold steady in
the face of various means of achieving them. And finally, we assume that decision makers are
utility maximizers. That is, decision makers will select the alternative option that gives the
greatest amount of net benefits, or the greatest benefit at the lowest cost. And while we're making
these assumptions, it's also important to notice what we're not assuming. Note that the rational
actor model does not assume that every decision automatically results in a sound outcome. Instead,
the rational actor model holds that the rational decision-making process will lead to better
decisions but not necessarily better outcomes. Now that we have a basic framework for
understanding rationality, let's think about the rational actor model itself. The rational
actor model is commonly assumed to be a realistic description of the decision-making process--or
at least the idealized decision-making process decisions--result from a rational and optimal
process under the decision-maker's control. It's perceived to be consistent with the formal
organizational structures and hierarchies that exist in most bureaucracies. But it rests on
several key elements. First, it assumes that the executive branch operates according to a pyramid
of authority, with the president at the top, and exercising power over foreign policy. Second, it
assumes that advice and information from advisors and from the bureaucracy flow to the president,
and that the president makes choices based on this advice and information. Further, it paints the
president and the staff within the White House and the cabinet as able to coordinate the bureaucracy,
and it paints the bureaucracy as responsive to the president. We'll consider the degree to which all
of these are true in other videos. Now even with all of those assumptions, the rational actor model
is probably the way we most often assume policy is made. The model itself specifies several steps in
the decision making process. First, a problem is identified. Let's say in this case we're concerned
about the North Korean nuclear program. Ideally, we'd have full information about the actors
involved, their motivations, their capabilities, and the nature of the international
environment or the context of the problem. In the case of the North Korean nuclear program,
the rational actor model would assume that we understand the scope of the North Korean nuclear
program, the country's motivation for pursuing nuclear weapons, and we have an understanding of
the broader international context. That is, what's China's position? South Korea's position? And so
on. Once policymakers have identified the problem, they then outline their goals, identifying and
ranking their preferred outcomes to a given situation. With respect to a particular problem
in this case, maybe the preferred outcome for the United States would be the denuclearization
of North Korea. But it might be willing to accept lesser outcomes as well, maybe a limit
on the number of nuclear weapons held by North Korea or some other set of outcomes. Next,
decision makers will gather information about the problem and determine what options might
be available to them to achieve their goals. In our hypothetical example, policymakers may
consider a number of policy options. Maybe they consider diplomatic negotiations with North
Korea, or economic sanctions, or a cyber attack, limited airstrikes, a full-scale military
invasion, or some other set of options. Then the various alternatives are evaluated.
Policymakers will consider the potential consequences and the effectiveness--or the
cost and benefits--of each alternative, and the probability that each will lead to
the preferred outcome. That is. to success. They may determine, for example, that North Korea
would respond with a nuclear attack if the United States launched an airstrike or a full-scale
military invasion against the country. The cost of such action would therefore be very high.
But it might also conclude that the effectiveness of diplomacy was limited, given the successive
rounds of negotiations with North Korea that have already occurred. Thus, while the cost of such
action might be low, so too is its effectiveness. Once all the alternatives are evaluated, the
decision makers will select the option that achieves the preferred outcome at the lowest
cost. That is, it has the greatest benefit--or the greatest chance of achieving the goal--at
the lowest cost. Let's say that the United States decides that expanding economic sanctions
on North Korea is most likely to achieve its preferred policy outcome at the lowest cost. That
option is then implemented, and the problem is monitored and evaluated to see if the implemented
solution is effective in achieving the goal. If necessary, additional problems are identified
and the cycle begins again. The rational actor model makes several assumptions in order to model
the decision-making process, and in doing so, it misses out on some of the context that's important
for understanding how decisions are actually made. The rational actor model generally assumes a small
number of like-minded decision-makers arriving at a consensus position. In reality, there are often
multiple conflicting power centers, each pursuing their own objectives. The rational actor model
generally assumes that states' preferences are unified and clearly specified. That is, a single
idea of the national interest exists, and decision makers operate in pursuit of that interest. In
reality, states' preferences are often contested. The rational actor model generally assumes that
there's a high degree of certainty in our goals and in the avenues available to pursue them. In
reality, decision making is often characterized by incomplete or even inaccurate information. The
rational actor model generally assumes that the implementation of decisions is done
in a relatively efficient manner. In reality, implementation is often subject to
its own bureaucratic and political processes, and can be delayed characterized by conflict,
miscommunication, and so on. And finally, the rational actor model usually views policy as the
optimal outcome of the decision-making process. That, is it views the policy outcomes that result
from the process as the best available option in pursuit of the specific goal. In reality, policies
can often be sub-optimal, ineffective, sometimes even contradictory or self-defeating. Despite all
its limits and shortcomings, the rational actor model is a useful way for thinking about the
idealized decision-making process, recognizing that it's a model and a theoretical framework
for understanding a process rather than an actual description of decision making on the ground.
We examine some of the challenges to the model in other videos. But that concludes our brief
overview. I hope you found it useful, and please leave any questions you have in the comments
section below. Thanks for watching everyone!