Leanne C. Powner

Publications, Papers, and Research
The Dissertation

Teaching and Course Materials

Summer Academic Working Group

External Links

FAQ

Personal

Meeting at Munich: Bargaining and Information in World Politics

Introduction
This simulation recreates the bargaining situation at the 1938 Munich conference, where Chamberlain "appeased" Hitler by giving him a region of Czechoslovakia known as the Sudetenland. Teams of students represent various states involved in the negotiation. They use two sets of documents - public information in the form of news article excerpts from the Times of London, and private information excerpted from each country's diplomatic documents of the period - as the basis for their bargaining. Because each side has some private information about willingness to fight, capabilities, or true preferences, their estimates of the others' likely bargaining behavior differ radically. As a result, instructors can emphasize the role of information and perceptions in bargaining, and in particular how this can make war occur even though all parties know that war is ex post inefficient.

Originally Intended Use
This simulation was originally conceptualized as a way to introduce students to the logic of Fearon's "rationalist explanations for war" (International Organization, 1995). In the way I normally use it, play proceeds as follows. I pre-establish country teams for Germany, Czechoslovakia, the United Kingdom, and France, along with a team for the United States and one for "others" - the USSR and Poland, primarily. The latter three countries were not actually involved in the negotiations at Munich, but they plausibly could have been; insufficient information is available in English to create an Italian team. In their country teams, the students estimate each country's ideal point, reservation point, and cost for war, beginning first with Czechoslovakia and Germany, and then repeating the process for each additional country that enters the negotiations. Students are not bound to follow the negotiation process as it originally unfolded, though most choose to do so. Some have sent their foreign ministers (another team member) rather than the leaders negotiating directly; once, a foreign minister was recalled and fired for accepting something outside the team's specified zone without consulting home first. Some groups have invited the US or France to mediate instead of the British, and a German group tried to bring the USSR in on its side to balance the British by promising it part of Poland. On one occasion, the Poles inserted themselves by threatening to seize unilaterally an ethnically Polish region of Czechoslovakia. Since the game is zero-sum, I typically draw a line on the board representing the percentage of disputed territory held by each, and give each side a large brightly-colored magnet to place on the line to represent its current bargaining offer. This allows both sides and the audience (uninvolved participants) to keep track of the negotiations, and it also visually clarifies for the participants how the zero sum model works. This helps substantially when we get to the Section Guide below.

The key point here, though, is that without the intervention of a third party, and in particular a third party who is willing to supplant the Czech preferences entirely with its own, war is almost inevitable. The private information held by both sides typically leads them to believe that no mutually agreeable solution is possible. Despite prior exposure to Fearon's main claims in lecture, most groups fail to realize that the costs of war will make a non-negotiated solution inefficient. In particular, Czech teams often fail to realize that their utility from retaining territory will be offset by the costs of fighting for that territory (i.e., that the costs and benefits on different dimensions are exchangeable in some common currency like 'utils'). German teams, in contrast, realize about half the time that they'd be better to moderate their demands than to fight for all of the disputed territory. Uncertainty about the outcome of war (see below) also contributes to this situation. A few groups have tried to resolve the crisis by introducing a second issue dimension. The Germans in particular frequently follow history and demand a referendum (plebiscite) in the disputed areas outside of what the Czechs are willing to cede. Czech teams have tried to bifurcate the issue into legal control of the disputed areas and treatment of the minority, where they are willing to offer concessions on the latter but not the former.

About half to two-thirds of the time, the parties reach some type of settlement with the help of a third (or fourth) party. The rest of the time, though, the parties are unable to agree and one side will declare war on the other. (They are informed of this option at the start of the negotiating; both sides have initiated the use of force at about equal rates.) If war results, I typically put two colors of poker chips (in a ratio that represents each side's relative power) in a bag and draw five chips to determine the war's outcome. Each chip represents 20% of the desired outcome; final values for fighting the war and accepting its outcome are calculated on this basis. Example: The outcome from war is 80% of the territory goes to Germany (four German tokens were drawn), and 20% goes to the Czechs. Each side then subtracts their previously determined cost of war from their war outcome to determine their final utility. Any countries that have entered the bargaining on behalf of a certain country are also considered part of the war coalition for determining relative power and final utilities. You may wish to have a pocket calculator handy at this point. To encourage in-character play, I offer teams bonus points - if a team negotiates a solution that gives them 80% of the disputed territory, then each member will each receive 8 (of 10 possible) bonus points. Members of the opposing team(s) will each receive two. Any losses from war are also converted into fractional points and are deducted from the bonus points. Providing bonus points also makes other teams eager to offer their "good offices" and get involved in the crisis.

After the sides have calculated their outcomes, we step back and pool information using the "Section Guide" worksheet (included below in the Full download). I normally do this on the overhead projector so that I can layer or change sheets easily and use multiple colors. One side's 0-100 axis is on the top of the line, and the other's is on the bottom. Since the situation is zero-sum, one side's ideal point (100) is the other side's 0. We establish this information first and determine the location of the status quo. Who believed what? The Czechs and Germans report their belief about the other's reservation point and cost of war, as well as their estimates of their own reservation points and war costs. (Using multiple colors of chalk/overhead pen is valuable at this point.) This establishes from the start that the two sides' beliefs do not match. It is often quite obvious at this point that one side - usually the Czechs - does not perceive a negotiated solution as at all possible (i.e., Czech estimates of the German acceptable-set [range of acceptable outcomes between the ideal point and reservation point] do not overlap with their estimates of their own acceptable-set). When no overlap between the acceptable sets - which is known as a win-set - exists, war is the likely outcome. This is true even if the other country's own 'true' acceptable set does overlap. (You can show this by putting the Germans' own preferences on a separate slide and overlaying it on the "Czech preferences and perceptions" slide.) One side's belief that no acceptable solutions exist is enough to result in war.

Then, we typically start on a clean slide with the French, and show that their estimates of German positions and their own position do not typically overlap. The only state whose beliefs allow for a win-set is typically Britain. This is why replacing the Czechs with the British in the negotiations with Hitler produces a non-war outcome. (If France's or another state's acceptable set did overlap with that state's belief about the German acceptable set, then this state too would have been an acceptable interlocutor in the negotiations who also would have appeased Hitler.) As we address each country, we usually talk about what specific information that side had that influenced its estimates. Students are frequently surprised by what other sides "knew" that they did not, and this provides a valuable chance to talk about the role of intelligence, diplomacy, and communications technology in crisis situations.

Finally, lesson plans and time allowing, we return to the Czech and German outcomes. If they negotiated an outcome, we discuss why fighting and then obtaining the same outcome was inefficient, and why this is inefficient even if one side obtains its ideal point. If negotiations collapsed and resulted in war, the war's outcome is usually very close to an offer made during the negotiations, and we revert to compare to that offer. (Note: You or a student should keep track of all offers made for this purpose.) Students then have a very graphic and memorable understanding of the effect of the costs of war. Imagine a bargained solution of 80% that would have given each member of the team 8 bonus points (see below). Imagine instead that the team fought and obtained an 80% outcome, but then had to pay the costs of war -- instead of 8 bonus points, they're now getting 8 minus their costs of war.  In some ways, then, having the negotiations break down is the best way to make the main point.

Plausible Variations
Plausible variations on this simulation could focus on the domestic politics which influenced all sides' ideal points and reservation values. The German, Czech, and British groups all have clear domestic pressures that influence their bargaining aims and strategies; in the German case, these pressures were manufactured explicitly for the purposes of boosting their bargaining position. As another angle, a substantial amount of signaling occurs in two different forms. First, we see direct A-to-B signaling, such as the Germans moving troops to the Czech border. We also see A-to-C-to-B patterns, where a third party is used to increase the credibility of a signal. In this particular case, we see several instances of German and Czech foreign ministers calling in ambassadors to reiterate their positions, and we also see Britain explicitly voicing the Czechs' own positions for them. This implies that the British are more credible as international speakers; they would not have agreed to make these statements and put their own reputations on the line if they did not believe the Czechs truly meant them. You might also assign multiple roles within country-teams: prime minister, head of state, foreign minister, chief of intelligence, minister of defense, etc., where intelligence and defense would be able to use the time between negotiating sessions (class days) to do additional research about the other sides' likely positions and capabilities. Other variations are possible as well, so please feel free to share if you develop a different angle.

Instructor Notes
As with all simulations, two steps are key to conducting them successfully. First, the instructor should think carefully about his/her goals in using the simulation. Are you teaching the Fearon model, as I was above, or are you more interested in something else? If your interest is emphasizing the domestic politics, allowing students the recourse to war if the bargaining fails is probably an unnecessary complication, but at the same time, you would probably want to supplement the documents in the Resource files below with some additional domestic politics material culled from similar sources. (I did not include much of this - for example, the Times' daily Labour responses to Chamberlain's handling of the crisis, the Czech debates on a plebiscite, or the major German rally and policy speech by Hitler - in the Resource files.) If you are focusing simply on the effect of uncertainty (private information) on bargaining, then introducing all the technical language and focusing explicitly on the costs of war is similarly an unnecessary complication.

Second, the instructor should think carefully about all the places that the simulation could go wrong. What if the parties somehow manage to settle without war and without introducing a third/fourth actor - how will you approach your main lesson then? What if the numbers don't work out when you try to talk about the models? What if one of the key players in the game turns out to be incompetent, lazy, out of character, or absent? What if no one has read the materials, or what if they've shared their private information with each other? What if the German team turns out to be mostly Jewish students who refuse to play? The more of these contingencies you identify prior to the simulation, the better equipped you will be to address them if they arise, and the more of them you may be able to avoid in your planning stages. Don't drive yourself crazy what-iffing, but do thoughtfully consider where the major junctions and critical outcomes are in the simulation for your particular set of goals and then make backup plans accordingly.

Additional resources for this simulation could include the case study in Carolyn Rhodes, Pivotal Decisions (2nd ed.), or another national newspaper (especially if you or the students have the ability to read or translate French or German). I believe that additional crisis-related documents from Soviet archives have also been published, though I don't have access to (or skills to read) them. If you do end up developing additional resources for this simulation, please share them with the rest of us. If you email them to me, I can post them on the site with full credit to you. Simulations and other pedagogical resources are public goods - please don't free ride.

Assessing the Simulation
In its original form, the simulation itself was not designed to produce assessable outcomes. Its purpose was to place students in the heart of a complex bargaining situation where all sides were acting strategically on the basis of their private information. As a result of that experience, students would have an intuitive understanding of the sides' incentive patterns, and would be able to intuit the logic behind Fearon's model. We then as a group elaborated the specifics of applying the model in this case. In short, the case itself is nothing more than a vehicle for introducing the model and several other concepts that we address immediately after this. The day-and-a-half simulation gives me about 6-7 class periods' worth of material and a continuing example for students to reference. By approaching the same crisis, which they now understand intimately, from a number of different angles and with a number of different emphases, they are able to triangulate their understanding and relate concepts to one another. The simulation itself is not intended to produce assessable outcomes, and it should not be used in its current form for that purpose. The assessable outcomes of this simulation are indirect and relate to the concepts the students experienced, rather than to the experience itself.

Downloads
NOTE: The private information files for the United States are not part of these documents. The US information comes from a volume of Foreign Relations of the United States, a State Department publication. The US documents contain a wide range of issues and angles; instructors are better off to select documents themselves which meet their own purposes, rather than using a set assembled for my particular purposes. FRUS volumes are available in most college or university libraries, even non-depository ones. If you don't have them on site, your reference librarian can ILL them very easily.

Full Simulation (*.zip) - All simulation documents, handouts, and resources. Student handouts are in MS Word '03 format (*.doc); main resources (newspaper articles and country private information sets) are in *.pdf.

Main Resources only (*.zip) - Public and private information sets; all documents in *.pdf.

Alternate document formats available on request.