International Studies Paper(Graduate students)

IAFF 6101 Interna-onal Affairs Cornerstone

Lecture 6 Coercion

October 4, 2016

Today’s Class •  Defining Coercion

•  Strategies of coercion

•  Coercive threats: the problem of credibility

•  Military compellence: finding the Achilles Heel

•  The sancFons debate

•  Smart sancFons?

Coercion

•  AHempFng to influence the behavior of another actor through the manipulaFon of costs and benefits – Costs: military and economic – Benefits: lots of things

•  Coercion involves the use of threats and reassurances – Threats: promises to inflict some kind of cost in the case of noncompliance

– Reassurances: promises not to inflict costs (or stop inflicFng costs) in the case of compliance

Two Forms of Coercion

•  Deterrence: Dissuading an adversary from taking an acFon (i.e., to not do something) by threatening him/her with undesirable consequences – Defense is about stopping an aHack by direct military resistance

– Deterrence is about persuading an adversary not to aHack in the first place

Two Forms of Coercion

•  Compellence: Persuading an adversary to stop an ongoing acFon or to start a new course of acFon (i.e., to do something) via the threat or use of force – Conquest is about completely defeaFng an adversary or taking what you want by brute force – with liHle or no cooperaFon from the adversary

– Compellence is about geVng an adversary to give you what you want without having to pry it from his cold, dead fingers •  It’s about achieving goals “on the cheap”

In Peace and in War

•  Both deterrence and compellence can occur in peaceFme and in warFme

Mapping Coercion Peace-me War-me

Deterrence Deterrent threats Intrawar deterrence •  U.S. threat vs. Iraq in 1991 not

to use CW •  Early WW2: “If you bomb my

ciFes I’ll bomb yours”

Compellence Compellent threats (a.k.a. coercive diplomacy) Threats or use of economic sancFons

Military compellence •  Blockade/sancFons •  Strategic bombing •  Limited nuclear use

Deterrence vs. Compellence •  Why is deterrence thought to be harder than compellence? –  Complying with compellent threat involves a recognizable submission to recognizable threat

–  Can damage reputaFon, lead to further challenges from same challenger or others

–  Complying with deterrent threat involves doing nothing; can always claim you never intended to aHack Country X

•  Challengers in compellent scenarios thus have incenFves to make it less embarrassing to comply –  Help the adversary concede

•  Cuban Missile Crisis

Deterrence vs. Compellence: Evidence

Source Success Rate Blechman and Kaplan 18% Petersen 24% George 29% Art 25% Art (expanded) 31% Sechser 41% Average Compellence 35% Deterrence 57%

The Coercion Calculus

V = B × P(B) – C × P(C) •  V Value of aHacking/resisFng •  B Expected benefits •  P(B) Probability of aHaining those benefits •  C Expected costs •  P(C) Probability of incurring costs

•  Coercion succeeds when V < 0

B •  V = B × P(B) – C × P(C)

•  The more the adversary values taking acFon/resisFng, the harder it will be to coerce

•  B is extremely difficult to influence but highly consequenFal –  Disagreement over Soviet moFves and objecFves led to

disagreements over U.S./NATO’s ability to deter –  Link to IR theory: greedy states value expansion more than security

seekers, will be harder to deter –  You oqen don’t know which type you face – uncertainty –  Worry about rogue states – do they value territory or inflicFng

damage so much that they are very hard (or impossible) to deter?

P(B) è Denial

•  V = B × P(B) – C × P(C)

•  Reduce probability that enemy will reap benefits of aHack or resistance

•  Goal = lower the likelihood that target’s military strategy will succeed

•  Deterrence by defense/denial: maintain capabiliFes to defeat adversary’s aHack

•  Compellence by denial: threaten/aHack military forces, or ability to transport or supply those forces

C è Punishment

•  V = B × P(B) – C × P(C)

•  TradiFonal punishment –  Increase costs to civilians –  Deterrence: threaten to inflict costs in retaliaFon for an aHack

•  Nukes –  Compellence: (threaten to) raise costs of resistance by inflicFng pain

on civilians, induce them to demand surrender –  Target set: things that affect civilian life and well-being

•  New (“smart”) punishment –  Increase costs to elites –  (Threaten to) inflict pain on leader or regime supporters, induce them

to change their mind or get rid of the leader –  Target set: things that affect well-being of elites

P(C) è Manipula-on of Risk •  V = B × P(B) – C × P(C)

•  Manipulate the probability that costs will be suffered

•  Goal = Inflict pain gradually; fear of future pain extracts concessions, not pain already suffered

•  Target set = same as Punishment

Credibility and Deterrence •  The lynchpin of deterrence is credibility – the opponent’s belief that you

will carry out the threat

•  Credibility is a funcFon of capability + willingness

•  Capability if is a funcFon of power; willingness is a funcFon of interests –  How much you care about the issue in dispute dictates the costs and risks

you’re willing to accept

•  If costs are high and interests are quesFonable, credibility is lower and deterrence more likely to fail –  Nuclear weapons – especially when you are vulnerable to retaliaFon (MAD) –  Extended deterrence – when you are defending a protégé against aggression

from a third state

•  U.S. faced both of these during the Cold War: nuclear threats to deter Soviet aggression in W. Europe

The Art of Commitment

•  Schelling: about “threats that are hard to make, the ones that are not inherently so credible that they can be taken for granted, the ones that commit a country to an acFon that it might in somebody’s judgment prefer not to take” (1966, 36).

•  Threats to hurt somebody can be credible even if they hurt you, too, if you can make them believe that you’ll actually do it

•  $64,000 quesFon = how to make them believe it?

The Art of Commitment

•  The raFonality of irraFonality

•  Relinquish the iniFaFve

•  Incur the naFon’s honor, reputaFon

•  Interdependence of commitments

Credibility and Compellent Threats (a.k.a., Coercive Diplomacy)

•  Get target to change its behavior by threatening to use force, or by using force in limited amounts (demonstraFve force) – CD fails if war/large-scale force is needed

•  What makes compellent threats credible? – Power?

•  Can be counterproducFve – why? –  Interests?

•  Usually favors the side defending the status quo – Regime type?

•  DemocraFc threats more credible?

Factors Associated with Compellent Threat Success

Variable Effect on Threat Success Water barrier + Distance + ConFguity - Challenger has long-range weapons - History of challenger-target conflict - Challenger’s capabili-es - Target has recently backed down + Militarized signal + Challenger is a democracy n.s. Challenger is a coaliFon - Threat is over leadership + Source: Todd S. Sechser, “ReputaFons and Signaling in Crisis Bargaining,” Journal of Conflict Resolu3on (forthcoming 2016).

Military Compellence

•  Compellence that happens during ongoing war

•  Key: find a short cut to victory; persuade adversary that resistance is fuFle or not worth the costs

•  Focus on strategic bombing

P(B) è Denial

•  Reduce probability that enemy will reap benefits of resistance

•  Goal = lower the likelihood that target’s military strategy will succeed

•  Target Set: –  DestrucFon of arms manufacturing –  InterdicFon of supplies –  DisrupFon of movement and communicaFon in baHle area –  AHriFon of fielded forces

Denial Strategies

•  Strategic InterdicFon –  Destroy enemy war producFon

•  ParFcular weapon •  CriFcal component •  TransportaFon

–  Works best in long wars

Denial Strategies

•  OperaFonal InterdicFon –  Prevent movement of forces in theater –  Prevent reinforcements from reaching theater –  Works best when front is fluid

Denial Strategies

•  Close Air Support –  AHack front-line forces, reinforcements

–  Works best with staFc front

Denial Works Best, But Not All the Time

•  ConvenFonal > guerrilla war –  Rolling Thunder (1965-68) vs. Linebacker I (1972) –  “Highway of Death” (1991)

•  Benefits of surrender > costs of surrender – NaFonal exterminaFon – Leaders fear punishment by public if they surrender

C è Punishment

•  Increase costs of resistance to civilians

•  Goal = INFLICT PAIN ON CIVILIANS, induce them to demand surrender

•  Target Set = urban areas, civilian infrastructure, electricity

Prophets of Airpower

•  Douhet, Command of the Air, 1923 –  Bomb civilians to induce widespread panic, societal collapse

•  RAF develops a “bomber culture” under Trenchard in the interwar period –  Moral effect of bombing is to material effect as 20 is to 1

Why Punishment Doesn’t Work

•  States accept high costs for important goals

•  ConvenFonal bombing can’t kill enough people

•  States can minimize vulnerability, adjust

•  Punishment doesn’t turn the popula3on against the government

P(C) è Manipula-on of Risk •  Manipulate the probability that costs will be suffered

•  Goal = Inflict pain gradually; fear of future pain extracts concessions, not pain already suffered

•  Target set = same as Punishment

Why Risk Doesn’t Work

•  Has all the problems of punishment, plus – – Can’t inflict enough pain – Hard to implement – Stops and starts communicate irresoluFon

Decapita-on è ??

•  Goal – Leadership – Kill enemy leadership (B? C?) – Poli-cal – Foment a coup (B? C?) – Military – Cut off leadership from fielded forces (P(B))

•  Target Set – Leadership – Enemy leaders – Poli-cal – Regime’s repressive apparatus – Military – Means of communicaFng with the military

Why Decapita-on Fails

LEADERSHIP DECAP – Hard to find leaders, killing them doesn’t always change policy, will successor be any beHer?

POLITICAL DECAP – Airpower ≠ good tool for fomenFng coups, coups hardly ever happen in warFme, coups need support of army

MILITARY DECAP – Hard to cut off communicaFons, lots of ways to communicate, pre-delegaFon

Military Compellence (Conven-onal): Summary of Findings

•  Punishment (of populaFon) doesn’t work

•  Risk doesn’t work

•  Denial can work, but success is sFll far from assured – Threatens target’s strategy of resistance – But depends on circumstances

Between Deterrence and Compellence, between Peace and War

•  Crises with nuclear weapons = compeFFons in risk taking

•  Nobody is likely to start a nuclear war on purpose, but a nuclear exchange could sFll happen –  “A response that carries some risk of war can be plausible, even reasonable, at a Fme when a final, ulFmate decision to have a general war would be implausible or unreasonable” (97).

–  Wars start via a process of escalaFon, not a bolt from the blue

•  Brinksmanship = manipula-ng shared risk of war –  “The risk of disaster becomes a manipulaFve element in the situaFon. It can be exploited to inFmidate” (102).

–  Persuade the adversary to back down

The Role of Nuclear Weapons in Coercive Bargaining

•  Nukes should not be judged on the basis of their baHlefield uFlity – it’s how they affect the risk of general war

•  NATO does not have to be able to win a local war •  Don’t need really strong convenFonal forces •  Don’t delegate launch authority – want strict centralized

control •  Plan for a war of nerve, not a tacFcal war – choose targets

less for military than for symbolic/demonstraFve effect •  SelecFve use rather than large-scale tacFcal use •  Nukes = signal; don’t want to wait to use them unFl you’re

desperate

Economic Sanc-ons (Tradi-onal)

•  Induce change in target’s poliFcal behavior by inflicFng economic pain via reducFon or restricFon of internaFonal trade or investment with target

•  Mild form of punishment

Hu`auer, Schoa, And Ellioa

•  Economic Sanc3ons Reconsidered

•  115 cases of economic sancFons, 1914-1990

•  40 successes: 34%

•  More successful than previously believed

Pape Fires Back

•  On closer examinaFon, only 5 successes, not 40

•  HSE failed to control for threat/use of military force

•  Modern states are tough nuts to crack

Selec-on Effects •  SancFons impose costs on both sender and target; both sides

have an incenFve to avoid them

•  Targets inclined to acquiesce will do so to the threat of sancFons

•  SancFons only imposed when target is highly resolved – and will fail

•  Most studies of coercion (including Pape’s) don’t get this, and thus seriously underesFmate efficacy of sancFons

Evidence of Selec-on Bias •  Drezner looks at cases where sancFons were threatened as well as imposed – and force is not an opFon – Threat/use of sancFons in U.S. trade disputes

•  Overall, DD finds the threat of sancFons was much more effecFve (66.7%) than imposiFon of sancFons (41.7%)

•  Argues that this is evidence of selecFon bias

Selec-on Bias in Militarized Compellence?

•  Drezner looks only at trade disputes – use of force not usually a possibility

•  How could we test for selecFon effects in disputes where force is possible?

•  Coercive diplomacy and compellent threats

•  Already saw that threats of force are not more effecFve than use of force to compel!

•  No evidence of selecFon bias in military sphere

Rise of “Smart” Sanc-ons •  Most sancFons unFl 2000 were “comprehensive” sancFons

•  Turning point = Iraq sancFons in 1990s

•  Enormous impact: 50% reducFon in Iraqi GDP…

•  …but primarily affected populaFon, not leaders – Huge effect on infant mortality – Largely conceded by Madeleine Albright

Target Elites •  Targets almost always authoritarian regimes – can divert effects of

generalized sancFons away from regime supporters onto average people

•  Need to target sancFons at leaders, core elite supporters of regime

•  Financial sancFons, asset freezes, travel bans, restrict luxury goods, arms embargoes

•  Pleases everyone! –  Countries in UN get to cooperate with hegemon –  No excessive humanitarian costs –  More precise, targets the real evil-doers! –  BiparFsan support in U.S.

Effect of Smart Sanc-ons •  Humanitarian effect = good

–  Less harmful than comprehensive sancFons –  But comprehensive sancFons seem to work beHer someFmes

•  Target = democracy •  Goal = regime change •  For ending civil wars

•  Less effecFve at gaining target compliance –  Arms embargoes succeed only 8% of Fme –  Financial sancFons not effecFve for higher stakes –  Economic effect = unpredictable –  Only big success was Libya in 2003 –  Iran?