The Drone Dilemma: Veto Players, Moral Hazard and Decision-Making

My paper seeks to address the following question: does the current decision-making process to employ drones for military operations, in which the executive exercises unilateral control and informs the legislative branch after the fact (ex posto facto) create moral hazard(s)? Further, it seeks to address a question that has been almost entirely ignored thus far: what are the plausible alternatives, and would such alternatives also create moral hazard(s)? I define moral hazard as any situation in which an agent has a perverse incentive to take unnecessary risk because that agent does not incur the possible costs of failure and/or faces no repercussions for failure. Further, any situation in which an agent is preventing from performing a morally necessary action will be judged as a potential moral hazard. As a final prefatory note, I outline two characteristics of UAV combat that make it a particularly problematic case: swiftness of action, the inherent strategic benefit of acting quickly and decisively; and covertness, the inherent tension between the necessity of the Executive to conduct operations in secret and the Legislative’s duty to preserve democratic accountability. I conclude that the current unilateral decision-making process is indeed morally hazardous. However, I also conclude that plausible alternative schemes such as the necessity of legislative or judicial authorization before the fact (a priori) also create moral hazards. There is no right answer, only varying schemes which appease the concerns of those more interested in the effective conduction of security policy via unilateral control on the one hand; and those more interested in preserving democratic accountability on the other. I urge the reader to note the many questions this paper does not seek to address, including the legality of targeting American citizens—this is another subject entirely.

To test my inferences I employ an alteration of Professor George Tsebelis’ Veto player theory (an abridged explanation of which can be found in my first blog post). For three different systems, I employ this game theory to exemplify how the normative institutional structure of the decision-making process will affect the behavior of internal political actors. The easiest way to detail the process is to simply provide the figures and their respective potential moral hazards (PMHs):

Ex Posto Facto Congressional Review (Current System)


PMH 1.1: Lack of Accountability

PMH 1.2: Perverse Incentive to Use Force

PMH 1.3: Politicization of Force

A Priori Congressional Approval (Hypothetical Alternative #1)


PMH 2.1: Prevention of Morally Necessary Action (via the increased likelihood of inaction due to institutional constraints)

PMH 2.2: Politicization of Force

PMH 2.3: Potential for Intelligence Leaks

PMH 2.4: Sacrifice of Legitimate Congressional Review

A Priori Judicial Authorization (Hypothetical Alternative #2)


PMH 3.1: Prevention of Morally Necessary Action

PMH 3.2: Perverse Incentive for the Executive to Disinform

The paper details each PMH but space prevents me from doing so here. If you’re interested I’ll be happy to discuss my conclusions in more detail at the conference. See you soon!

— Miles Kellerman (University of Wisconsin)

This entry was posted in 2012-2013 General. Bookmark the permalink.

2 Responses to The Drone Dilemma: Veto Players, Moral Hazard and Decision-Making

  1. scotthiers says:

    I love your topic, Miles. And it could not have come at a more opportune time. I am looking forward to hearing about your conclusions in greater detail. What assumptions did you employ in your game theoretic model? I have not had a chance to read the foundational veto player article, but I am guessing you had to assume risk neutrality on the part of the executive and risk aversion for Congress?

    Be prepared for plenty of false comparisons to the current domestic drone use issue, as I am sure they will abound at the conference (even with your forewarning).

  2. mkellerman says:

    Thanks Scott. I adopted the assumptions of Veto Player Theory as proposed by George Tsebelis:

    His theory is about comparative institutions (such as Parliamentarianism versus Federalism, simple versus supermajority voting, etc.). Integrally, he focuses on how comparative institutions affect the possibility of changing existing policy, which he terms the ‘status quo.’ There are a number of assumptions about the players, including transitive preferences and information symmetry, though it varies depending on the type of institution. The idea is not to predict what will happen—this would be extremely difficult given the unpredictability of individuals, but rather how much easier or harder the institutional design makes changing existing policy. The larger the Winset (points that defeat the status quo given that they satisfy all parties), the easier it is to come to agreement and initiate a change; the larger the core (the points that do not satisfy enough veto players to create agreement) the less likely.

    I added on major alteration: in the depiction of the current system, I use dashed lines to represent a player that reviews legislative after the fact. Because ex posto facto review is an extremely weak counter to the Executive’s decision-making calculation (especially when that Executive has a tremendous information advantage over the reviewing body and thus could potentially distort the details) it does not affect the Winset—the points that can defeat the ‘status quo,’ or, in the context of this paper, the points which would satisfy all veto players and allow for the initiation of the policy.

    I use a very simplified version, and I’m using the theory in pretty unorthodox way by replacing a literal instituions like a parliament with a normative institution—the decision-making scheme to use drones. There are an enormous amount of articles that apply veto player theory to different subjects, and there are also some legitimate criticisms of its assumptions. But I think it works well as a detail of how different decision-making procedures *could* potentially affect the process. Because I’m not predicting what will happen, I use *potential* moral hazards rather than moral hazards.

    Phew, long-winded, sorry!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s