Principal Agent Theory and Artificial Information Asymmetry

Published date01 June 2009
Date01 June 2009
DOI10.1111/j.1467-9256.2009.01349.x
Subject MatterResearch and Analysis
Principal Agent Theory and Artificial Information Asymmetry





P O L I T I C S : 2 0 0 9 V O L 2 9 ( 2 ) , 1 3 7 – 1 4 4
Research and Analysis
Principal Agent Theory and Artificial
Information Asymmetry

Stephan Poth and Torsten J. Selck
University of Nottingham
This article evaluates the potential for the concept of ‘artificial information asymmetry’ to be used
in political studies which focus on principal agent theory. We present real-world examples and
excerpts from political writings to support our argument that, although artificial information
asymmetry appears to be an important factor in principal agent relations, the concept has been
underemployed in the political science agency literature. To be able to present more realistic
scenarios of delegation, political scientists might be well advised to incorporate the concept so that
it can be used in applied research.
Introduction
In 1587, Mary Stuart was finally executed for treason after a long imprisonment.
Her rival to the British throne, Elizabeth I, had signed the death warrant but had
hesitated to order the execution. Elizabeth’s council saw it as its duty to, ‘in
apparent defiance of the Queen’s will, do what she really wanted and would not
expressly command’ (Zweig, 1935, p. 333). Whereas the tragedy of a lifelong
passionate rivalry and two decades of domestic political insecurity ends here, the
comedy of trying to shift responsibility reaches its climax when Elizabeth bemoans
a disastrous error and scandalous plot, a ‘usurpation of power by subordinate
authorities’ (ibid.). Although straining the truth, this shirking of responsibility was
expected to serve two main objectives. Firstly, the pretence should preserve Eliza-
beth’s moral prestige as a monarch among her European peers. Regicide was
unthinkable because it undermined royal sanctity. Secondly, it might have had a
psychological effect as it helped to silence her conscience.
A rather striking anecdote in Machiavelli’s The Prince has certain underlying paral-
lels with the events mentioned above. Cesare Borgia trusted the cruel efficiency of
his Spanish minister Remirro de Orco to ‘pacify’ a newly secured territory and
‘make it obedient to the sovereign authority’ (Machiavelli, 1999, p. 25). After
successfully fulfilling his task in the anticipated manner of excessive authority he
was found cut into two pieces in a public place. Machiavelli outlines Cesare Borgia’s
motivation for the murder:
‘Knowing also that the severities of the past had earned him [Cesare Borgia] a
certain amount of hatred, to purge the minds of the people and to win them over
© 2009 The Authors. Journal compilation © 2009 Political Studies Association

138
S T E P H A N P O T H A N D T O R S T E N J . S E L C K
completely he determined to show that if cruelties had been inflicted they were not his
doing but prompted by the harsh nature of his minister
’ (Machiavelli, 1999, p. 25,
emphasis added).
Here again, responsibility was shifted to an allegedly ‘uncontrollable’ agent running
riot. By making him a scapegoat, Cesare Borgia deflected attention away from his
own responsibility. The murder appeased the people who had developed hatred
against Cesare Borgia himself due to the harsh treatment in the past by his own
obedient and faithful agent.
The cases we have briefly summarised are not chosen simply for their tragic
attractiveness. These curious historical anecdotes offer a wealth of insight into the
phenomenon of artificial information asymmetry. We borrow the term artificial infor-
mation asymmetry
from Neil J. Mitchell, who used it in his study on atrocities
committed in civil wars based on conscious leadership decisions (2004, also quoting
Machiavelli’s account of Remirro de Orco). He is to our knowledge so far the only
one who applied it in the context of decision-making. The cases above cover
fundamental features of a vital aspect of principal agent theory. In both accounts we
observe leaders whose policy choices are constrained by some kind of ‘principal’.
Respect among peers and possibly her own conscience counterbalance Queen
Elizabeth’s personal and political rationale to rid herself of a dangerous trouble-
maker. As far as Machiavelli’s Cesare Borgia is concerned, this passage focuses on
the loyalty of his subjects and the avoidance of a revolt. In contemporary demo-
cratic polities we would add enfranchised citizens and their power of electoral
punishment. Nevertheless, the preferred policy is implemented by their agents. In
the absence of directly expressed orders, lines of accountability get lost in the fog of
principal–agent relationships, and principals go away unpunished and might even
find a convenient scapegoat in the form of the agents. ‘This political technique of
the principal, in which he knows but finds it convenient to conceal his knowledge,
in which he is in control but pretends not to be, might be called artificial informa-
tion asymmetry’ (Mitchell, 2004, p. 48).
Traditionally, the principal agent logic emphasises structural constraints on leader-
ship. We argue that the very same dynamics inherent in the model potentially
increase a principal’s options by giving him or her the means to shirk responsibility.
This article’s primary aim is to outline the strategy of artificial information asym-
metry. It seeks to encourage scrutiny of the analytical usefulness of Mitchell’s
concept with regard to the process of political decision-making in future research.
We believe that a reassessment and critique of the...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT