Theories of audience costs are a relatively new phenomenon within international relations, with Fearon (1994) arguing that democracies are able to generate audience costs and thereby are seen to be more credible when making threats on the international stage. This essay will argue that some autocracies can generate varying levels of audience costs, dependent upon the type of autocratic regime that they follow. In regimes where there is little personal connection between those peoples in the political elite and the leader, more audience costs can be generated as the elite have the ability to overthrow a leader without losing their own positions. However, in authoritarian regimes where there is a greater centralisation of power around a singular leader, and political elites rely on the leader to maintain their positions, it is more difficult to generate audience costs. Thus, using Fearon’s classical theory of audience costs, it would be possible to signal credibility through the generation of audience costs.

Then, through the analysis of the underlying assumptions of the classical model of audience costs, this essay will go on to argue that despite some autocratic states being able to generate audience costs, these do not particularly matter as a means of signalling credibility. It will be noted that there are other ways of signalling political credibility to international states such as creating a reputation for not bluffing. Furthermore, leaders in both democratic and autocratic states are able to mitigate the effects of audience costs through using techniques that involve retrospective behaviour to justify not following through with a threat. Due to the fact that audience cost theory does not necessarily hold in a democracy, even if autocracies are able to generate audience costs, it does not necessarily matter as a means of signalling credibility.

Defining classical theories of audience costs

One of the ways of making a threat credible is through using a technique known as “tying hands”. This involves a leader taking an action in which the costs of backing down on an international stage are increased in the instance that a challenge from the opposition is presented. If no challenge is presented by the opposition, however, there are no costs to the leader that took the action. A primary way of doing this is through creating audience costs. By making a visible and public statement of intent, a state leader is able to put their own credibility and their state’s credibility on the line. Domestic audiences are able to observe and assess the performance of a leader during an international crisis. If the leader is found to be indulging in “cheap talk”, making threats and then backing down, Fearon (1994) argues that the leader will suffer from costs which arise due to the concerned domestic audience, who worry whether the leadership is successful, or view the leader to be unsuccessful at foreign policy. It is argued by Fearon (1994) that the stronger domestic audience a leader has, the more credible the leader can be perceived in the international arena, as signals made by the leader become costly to his own political stability. Thus, democracies are more able to signal their intentions convincingly, and are less likely to indulge in “cheap talk” as their domestic political audience would punish them by removing them from office at the next election.

There are three central features of audience costs, according to Fearon (1994). The first is that in order for audience costs to exist, a domestic political audience is required which have the ways and means to co-ordinate and punish a leader. In autocracies, sanctioning a leader is thought to be riskier than in democracies as autocratic leaders are largely unaccountable to domestic groups (McGillivray and Smith, 2000: 815). However, most autocratic regimes do require the support of domestic elites who can act as audiences. Secondly, actors in the domestic sphere must view backing down after making a threat as less desirable than conceding prior to making a threat. Thirdly, the domestic possibility of sanctions for backing down must be observable to outsiders.

In democratic states, foreign policy is formulated by an agent who is representing an electorate; voters are the principal driver behind foreign policy as they have the power to electorally sanction the leader or deflate the public opinion of him. There is a contract between the electorate and the political representative, in which the electorate would want to punish the leader for inflating a crisis through making a threat, and then backing down. In this classical model of audience costs, it is assumed that in autocracies, the leader constructs foreign policy themselves and is not able to credibly commit to self-imposed punishment (Katz,1991: 307). This shows an underlying assumption that the leader himself would be the only person willing to sanction the leader in an autocracy would be the leader himself, and there is no domestic political audience who would do so. To determine that this is not how audience costs work in all autocratic states, it is imperative to assess autocratic regimes against these criteria in order to establish how some autocratic regimes are able to generate audience costs.

Audience costs in autocratic states

Whilst discussing audience costs with relation to autocracies, it must be noted that autocracies differ from regime to regime, and therefore the classical model of audience costs outlined above cannot be applied to each as a generalisation. Geddes (2003) groups autocratic regimes into three distinct types: single-party regimes, military regimes and personalist regimes. Although these regimes all fall under the umbrella term of “autocratic”, it is important to look at each regime individually to assess whether audience costs exist within them.

Single party regimes are where one party dominates the political system even though competitive elections are held for certain offices. High levels of power are not occupied by relatives of the leader; rather individuals who have worked their way up through party rankings in order to gain their importance (Weeks, 2009: 68) This independence of people in high office from the leader means these political elites are able to act as an audience for the leader, and would be secure in the fact that if they were to oust the leader, their positions within the regime would be secure. This structure of government is easily observable to foreign leaders, and thus leaders operating under a single-party regime are able to generate audience costs from the political elite. Under Fearon’s traditional theory of audience costs, these would work in order to signify a more credible threat on the international stage.

Military regimes are defined as being autocratic regimes which are governed by an officer or retired officer, with the support of military establishments and some routine mechanism by which high level officers could influence policy choice and appointments (Weeks, 2007). In this situation, the military rather than the population are able to act as an effective domestic audience for the leader, and would have the potential to punish a leader for any “cheap talk” made. Leaders of a military regime, therefore, are more vulnerable to being overthrown than leaders of other types of autocracy, and arguably, they face audience costs that are not too different from those faced in a democracy (Frantz, 2008: 84). This is due to the fact that there is a low centralisation of power around one supreme leader, thus making it more difficult to manipulate the electorate in order to diminish audience costs under Fearon’s original theory of audience costs.

The third type of autocracy defined is that of a personalist regime, in which there is a high centralisation of power around a leader. In this case, there is no domestic political audience that can effectively sanction the leader for engaging in “cheap talk”, thereby making the leader less credible on an international stage. A leader of a personalist regime has the ability to punish citizens involved in internal crises, in order to stop themselves being overthrown, through means such as imprisonment, or even death (Weeks, 2008: 97). The fate of the political elites within a country operating under a personalist regime is closely connected to the leader, and they can be disciplined accordingly, such as the fate suffered by those under the Pinochet regime. For elites in this situation, keeping a leader in office is more desirable than risking their own careers, and there is no credible threat of removal in the face of bluffing. This concentration of power is evident or assumed by foreign countries; there is no reason to believe that Kim Il Sung had any internal challenges to his North Korean regime.

Do audience costs matter in autocracies?

It does not necessarily matter whether audience costs do or do not exist in autocracies as there is little evidence in real life crises to show that they actually operate; majority of the audience cost literature focuses on simulated crises. Snyder and Borghard (2009) argue that the domestic audience within a country would act to save their country’s reputation for reasons of resolve and national honour independently of threats; it does not matter whether or not a leader is engaging in “cheap talk”. This shows that there is a difference between the way that theorists observe audience costs and the way that they actually drive domestic politics within a country. This potentially makes the literature on audience costs null and invalid as working in simulated crises is not enough to prove that they exist in real life crises.

It is imperative to note that audience costs, as described by Fearon, are certainly not operating in the way that is suggested, even within democratic nations. Democratic leaders are aware of the consequences of their actions, and therefore avoid making reckless public threats if they feel that they cannot follow through (Shultz, 2001: 23). Whilst there is evidence to show that democracies do act consistently with audience cost theory, it is impossible to determine whether audience costs are the reason behind the operation. There are also forces of selection effect evident within democracies, whereby democratic leaders are cautious and only issue threats that they believe will be pursuable, making it very difficult to observe occasions where leader are caught bluffing. In general, most threats made by democratic nations are only put forward only at times when the party in opposition agrees with the substance of the policy (Snyder and Borghand, 2009: 8). Thus, the threat is made in a way that suggests the democratic nation is unified behind a particular policy formation, and it is in the country’s power and interest to make a threat. Here, audience costs are considered minimal, as even if the electorate were to rid the leader of their power at the next election, it would not make sense for them to do so as the opposition party were in support of making the threat. This makes the use of empty threats an issue of national interest rather than an issue of credibility.

Even if audience costs, at face value, were to exist within a country, it is exceptionally easy for a leader to reduce the costs that are generated through this mechanism by claiming a change of circumstance. Committing to a threat induces a degree of interpretation, and there are numerous ways that a leader can reduce the costs of backing down. (Snyder and Deising, 1977: 84) For example, the leader could argue that the opponents actions were not of the type the threat intended to coerce, that the opponents concessions in unrelated issues mean that it is no longer required to carry out the threat, or that there is another way in which counteraction can place in the future. A leader could claim that as the crisis was not planned, a concession is necessary in order to improve overall relations with the opposition country, or that the retreat was forced by a domestic necessity rather than the threatened opponent (Snyder and Deising, 1977: 202). A skilled leader, whether hailing from a democracy or an autocracy, therefore, is able to alleviate the potential audience costs of “cheap talk” through a mixture of ambiguity and creativity whilst talking to their domestic addressees. For this reason, even if audience costs do exist within a nation, it does not necessarily mean that the public will act upon them if they are led to believe that a threat was not carried out for appropriate reasons.

It is also necessary to question the initial assumptions that are made in Fearon’s literature in relation to audience costs. In order for the theory to be effective, it is required that a public exists who are willing to punish the leader for making empty threats. Slantchev (2006) establishes that it is difficult to note when situations arise that require audience cost theory to be highlighted, as they rarely arise, particularly when a public favours soft line policy prior to a threat being made; this is known as policy costs rather than audience costs, where the public punishes a leader for not following the preferred policy, rather than making an empty threat. Even if a leader is able to mitigate audience costs through citing changed circumstances, policy cost theory suggests that the public would be relieved that the leader was not going to follow through on reckless commitments which they did not support. Shultz (2001) shows that in the Fashoda crisis, there was an inability to differentiate between audience costs and policy costs – whilst diplomats considered hands to be effectively tied, the public demanded strong policy action before the threat was made. Policy preferences during the crisis were established prior to tying hands, and it was the willingness of the public to support hard line domestic preferences that explain the commitment, not credibility through audience cost creation. Therefore, in this instance, audience cost literature is, in effect, made redundant.

Audience cost theory additionally suggests that the leader of the nation state must be conscious of domestic audience cost incentives before they make a threat to an opposition. The threat of disposal by the domestic audience may help the leader to prevail even if the leader’s actual purpose was to influence only on the international stage (Snyder and Borghard, 2009: 10). It is also necessary that the target understands the mechanism; the target must look for information about the actions of the threatener’s audience in order to assess the likeliness of audience costs and the vulnerability of the leader to them. If the target is not able to do so, regardless of whether it is a democracy or an autocracy, then audience cost theory does not hold, even if audience costs can be generated. It is also arguable as to whether the target state actually sees audience costs as a viable indicator of credibility, as they may just be basing their ideas on other indicators, such as reputation.

The domestic audience, whether in a democracy or an autocracy, must be aware if there is an inconsistency between the threat that a leader has made and the subsequent behaviour (Snyder and Borghard, 2007: 11) and care enough about the differences to outweigh any other considerations in order to punish the incumbent leader. Even though one does not generally associate autocracies and free medias, there are regimes which are not fully repressive and it can be difficult to observe from an external perspective the extent to which it is so, and the ability to generate audience costs is arguably explicitly dependent upon the media freedom from political interference within a country (Slantchev, 2006: 468) Therefore, even if a leader is able to generate audience costs dependent upon their regime, if there is not enough freedom of information being passed throughout the domestic audience, it may be unlikely that the domestic audience is aware or cares about the actions a leader takes on an international stage.

Conclusions

Autocracies are able to generate audience costs to an extent, dependent upon the regime that they follow and the extent to which power is centralised under a key authoritarian figure. Whilst it is not the general public who have the power to punish the leader for backing down after making a threat in an international arena, there are core elites within autocratic regimes who are able to act as a domestic audience. In a military regime, the military is able to act as a domestic audience, and it is this type of autocratic regime which is able to generate the most audience costs; in a personalist regime there is no domestic audience to punish the leader therefore the leader is able to say what they want on the international stage without fear of being denounced. Thus, if audience cost theory is to be used, it would be rational to assume that a threat from a leader of a military regime is more credible than a threat from a leader of a personalist regime.

It has been outlined that audience costs, however, do not particularly matter when assessing the credibility of a nation; the theory has proved to be ambiguous. It is rare that democracies, let alone autocracies, make explicit threats when a crises is at the brink of war, as leaders see them as reckless and irresponsible; a leader is able to mitigate the effects of audience costs if they back down through a variety of escape clauses. There is a discrepancy between policy cost theory and audience cost theory, and it is difficult to recognise if a punishment is for an unpopular policy or the failure to carry out a threat. In addition, reputation is cared about as an independent notion to the meeting of threats, showing that the very narrowly defined audience costs are merely a small part of the domestic context. Thus, even though autocracies are able to generate audience costs to an extent, they do not particularly matter on the international stage.

Bibliography

Fearon JD, (1994) Domestic Political Audience Costs and the Escalation of International Disputes, The American Science Review 88(3) p577-92

Geddes B (2003) Paradigms and Sandcastles: Theory Building and Research Design in Comparative Politics, Ann Arbor: University of Michigan Press

Frantz E (2008) Tying the Dictator’s Hands: Elite Coalitions in the Authoritarian Regimes, University of California, Los Angeles

McGillivray F, Smith A(2000) Trust and Cooperation Through Agent-Specific Punishments, International Organization 54 (4) p809-24

Katz M, (1991), Game Playing Agents: Unobservable Contracts as Precommitments, Rand Journal of Economics, 22:307-28

Schultz, K A (2001). Democracy and Coercive Diplomacy. New York: Cambridge University Press

Slantchev BL (2006) Politicians, the Media and Domestic Audience Costs, International Studies Quarterly 50: p445-78

Snyder J and Borghard E, (2009), The Cost of Empty Threats: A Penny, Not a Pound, American Political Science Review 105 p437-56

Snyder GH and Diesing P (1977) Conflict among Nations:  Bargaining, Decision Making, and System Structure in International Crises.  Princeton:  Princeton University Press

Weeks J (2007) Leaders, Accountability, and Foreign Policy in Non-Democracies, Paper presented at the 48^th Annual Convention of the International Studies Association, February-Match, Chicago

Weeks J (2008) Autocratic Audience Costs: Regime Type and Signalling Resolve, International Organisation 62(1), p65-101

___

 Written by: Kriti Bami
Written at: University of Leeds
Written for: Graeme Davies
Date written: January 2012

Further Reading on E-International Relations

• Cyber War Forthcoming: “It Is Not a Matter of If, It Is a Matter of When.”
• Trust Me If You Can: Voluntary Sustainability Programs in the Uranium Industry
• Do the Dead Matter? Identifying Mourning and Funerals As a Political Act
• “Fake It Till You Make It?” Post-Coloniality and Consumer Culture in Africa
• ‘Keeping it Real’ or Keeping it Realist? Hip Hop in International Relations
• The Responsibility to Protect: A Disputed Matter