China's Territorial Disputes (6 page)

Crucial to the understanding of two-level games is the concept of the so-called win-set, or plainly, “bargaining space” or “area of compromise,” that is, a political space in which negotiators on both sides can at least satisfy their minimal mandates, defined as elements making up the national interest by both Level I state agencies and Level II domestic constituencies. Specifically, the win-set for a given Level II constituency is the set of all possible Level I agreements that would “win,” which technically means gaining the necessary majority among the constituents when simply voted up or down.
³
While Level I refers to what Putnam calls the chief of government (COG), or what I prefer to call the chief government negotiator, the actors at Level II may represent bureaucratic agencies, interest groups, social classes, or even public opinion,
⁴
groups of people who can veto, impede or expedite a ratification process. Ratification may entail a formal voting procedure at Level II, but Putnam uses the term to refer to any decision process at Level II that is required to endorse or implement a Level I agreement, whether formally or informally.
⁵
With territorial or boundary disputes, Level II ratification should be construed to mean legislative consent for both the boundary agreement and the start of on-the-spot demarcation works in preparation for territorial transfer. Indeed, Putnam recognizes that “expectations of rejection at Level II may abort negotiations at Level I without any formal action at Level II.”
⁶
This, as we shall see, turns out to be the case with talks aimed at resolving the recurring Tiaoyutai/Diaoyutai/Senkaku territorial dispute, the doomed border talks between China and India from 1959 to 1962, and between China and the Soviet Union in 1964 and from 1969 to 1978.

The requirement that any Level I agreement must, in the end, be subject to ratification at Level II imposes a crucial theoretical link between both levels.
⁷
The relative size of Level II win-sets thus assumes critical theoretical importance -by affecting the possibility and distribution of joint-gains from international bargaining. The two-level game framework theorizes that the larger the perceived win-set of a negotiator, the more flexible his negotiating position, and the more he can be pushed around by his counterpart sitting opposite at the negotiating table. On the other hand, the small win-set can be a negotiating advantage, for the negotiator can then credibly inform his counterpart that “I’d like to accept your proposal, but I could never get it accepted at home.” However, a win-set that is so small as to be effectively absent means that negotia

tions are almost certain to fail or be aborted at Level I at some point in time, since it is virtually impossible for any agreement to be ratified in this case. If negotiations should continue, they would certainly be regarded as nothing more than a ritual to keep a line of communications open on the part of both sides. Such was the case with the border negotiations between China and the Soviet Union from 1969 to 1978.

The theoretical implications of win-set size are summarized diagrammatically in Figure 2.1 below, representing a simple two-issue bargaining game between negotiators X and Y. The two negotiators are faced with tradeoffs across different issues: in this case, how much to yield on issue 1(territory) in order to get issue 2 (trade), and vice-versa. Like the two negotiators, various groups at Level II are likely to have different preferences on the several issues involved in a multi-issue negotiation. The implications of these tradeoffs for the respective win-sets can be analyzed in terms of “political indifference” curves, with the operating measure being the loss of political support or votes needed for ratification by the negotiators. This is analogous to conventional indifference-curve analysis, with Figure 2.1 providing an illustrative Edgeworth Box.

Figure 2.1
Effects of reducing win-set size for two-issue negotiations

The most preferred outcome for Y is the upper right-hand corner Ym, this being the outcome that wins unanimous approval from both territorial nationalists and trade advocates. Likewise, Xm represents then maximum outcome for X. Each curve concave to point Ym represents the locus of all possible tradeoffs between the interests of the territorial nationalists and trade advocates, such that the political support or vote in favor of ratification at Y’s level II remains constant. The contour Y11-Y21 represents the minimal political support or vote necessary for ratification by Y for both issues 1 and 2 at the minimalist position 1,and the wedge-shaped area northeast of Y11-Y21 represents Y’s win-set. Similarly, X11-X21 represents the outcomes that can be minimally ratified by X, and the lens-shaped area between X11-X21 and Y11-Y21 represents the set of feasible agreement. At this stage, it is possible for both parties to ratify any agreement within this range. If the win-set of Y were contracted to, say, Y12-Y22 (perhaps because a larger majority is required for ratification), outcomes between Y11-Y21 and Y12-Y22 would no longer be feasible, and if the theory bears out, the range of feasible agreements would be truncated in Y’s favor. However, if Y, emboldened by this success, were to reduce its win-set still further to Y13-Y23 (perhaps because unanimity is required for ratification), the negotiators would suddenly find themselves to be deadlocked, for the win-sets no longer overlap at all.

Given the importance of win-sets, we need to understand what circumstances affect win-set sizes. Putnam mentions three sets of factors:

1    Level II preferences and coalitions

2    Level II institutions

3    Level I negotiators’ strategies

The propositions or hypotheses underlying these three sets of factors as originally formulated or proposed by Putnam are not all amenable to testing or falsification. As such, I have reformulated, broken up, or recombined the various propositions of his two-level theory of bargaining in order that concrete observations can be derived which have a better chance of affirming, falsifying or more likely, modifying this relatively novel but useful theory in the literature of diplomatic bargaining.

Factor 1:the size of win-sets depends on Level II societal preferences and informal coalitions

The size of the win-set is directly proportionate to the cost of “no-agreement,” which is what the negotiator or groups within the state are willing to forgo to achieve an agreement. The cost of “no-agreement” reflects the relative attractiveness of maintaining the status quo in the face of a possible agreement to alter it. The higher the cost of “no-agreement,” the less attractive it is to maintain the status quo, hence the bigger the win-set for agreement; conversely, the lower the cost of “no-agreement,” the more attractive it is to maintain the status quo, hence the smaller the win-set for agreement.
⁸
The cost of “no-agreement” can be something like the increased likelihood of military conflict, disruption in economic ties, or territorial losses. Putnam describes constituents’ preferences as “homogeneous” if the costs or benefits of an agreement bears evenly on them, and “heterogeneous” if an agreement bears unevenly on them, in that only some constituents and not others get “hit.”
⁹
When the costs or benefits of a proposed agreement are relatively concentrated on certain segments of the population, that is, when constituents’ preferences are “heterogeneous,” it is reasonable to expect that the constituents whose interests are most affected will make a special effort to exert influence on the ratification process, if not on the negotiating process. Groups worried about the costs of an agreement often become politicized and organized, this reducing the effective win-set for negotiation or ratification or both. In the case of territorial boundary negotiations, there may be groups of people in the country who believe that, although a successful boundary agreement will reduce tensions and enhance relations between their country and the adversary, it will be achieved at their cost, for example, through the loss of grazing land or fishing grounds, and for the nationalistic-minded, the dispossession of the national patrimony and ruination of national dignity.

It stands to reason that, because it is much easier for politicized interest groups to organize and propagate their agenda in a representative democracy where freedom of opinion and organization is protected, it would be much easier for activists in these countries to act as political pressure groups both within and outside the government and the ruling party to create, manipulate and galvanize public opinion conducive to realizing their particular agenda. An argument thus presents itself. Negotiators and politicians can be more easily persuaded or pressured to sign or ratify an agreement, or desist from doing so, by calculations of political costs and benefits in a democratic country where diverse preferences are given play and taken into account, than in a non-democratic one where such preferences are not given expression. The corollary of such an argument is that it should be harder to resolve a dispute with a democratic country than with a non-democratic country, because there are more domestic constituents in a democratic country whose interests have to be satisfied, or at least not harmed, before they will find an agreement politically acceptable.

The size of the win-set also depends on the relative size of the “isolationist” forces, the less extreme “nationalist” forces who are at least suspicious of those who do not oppose international cooperation in general, and the “internationalist” groups who desire international linkages and multilateral arrangements.
¹⁰
Support for international agreements is theorized to be greater in more open economies that are trade-dependent, where for most of whose citizens the cost of “non-agreement” is high, as compared with more self-sufficient countries with low trade to gross national product ratios, where the cost of “non-agreement” is low. With all else being equal, more trade-dependent economies will have larger win-sets and be more willing to enter into agreements with other countries, while self-sufficient states will have smaller win-sets and drive harder bargains in the international agreements they make. For the purpose of determining trade-dependency, the absolute size of a country’s territory, population or economy should not matter. The absolute level of bilateral trade between two countries, on the other hand, should be highly correlated with the possibility of a successful outcome in negotiating disputes between them.

Factor 2: the size of win-sets depends on Level IIpolitical institutions

Ratification procedures in a country’s legislature, primarily the number of votes required for ratification, clearly affects the size of the win-set. If a two-thirds vote is required for ratification, the win-set will most certainly be smaller than if only a simple majority is required. However, Putnam recognizes that not all ratification procedures are formalized. He has pointed out that the Japanese propensity to seek the broadest possible domestic consensus before acting constrains their win-sets.
¹¹
This seems to be even more the case when opposition parties and recalcitrant elements within the governing party have to be brought over to the government’s side when the matter before the Diet is the ratification of a treaty, when a unanimous legislative opinion is desired. The theory predicts that, the stronger a state is in terms of the autonomy of central decision-makers from domestic pressure, in the sense that they have better control of the number of votes of the Level II constituency required for ratification, the weaker the relative bargaining power of its Level I negotiators. This hypothesis is supported by Helen Milner’s findings that a divided government decreases the chances of cooperation while improving the outcome of any agreement that is ratified from the perspective of the legislature’s preferences.
¹²

The logic of this argument leads to the testable proposition that a representative democracy with checks and balances in the government is offered more opportunity to advance its interests as a whole than would be possible for a “centralized undemocratic” state. This is when the chief negotiator of the democratic state can be much tougher in her negotiations by arguing that her fettered government and rowdy legislature will not accept all or even most of what is being asked of her by her negotiating counterpart. Even in the case of China and the Soviet Union, their diplomats could have claimed, and in reality did claim, that they were constrained by domestic political pressures. This claim may indeed be possible, but given the complicated and usually covert political bargaining that characterizes the politics of these regimes, it would be very difficult for outsiders or the opposite negotiator to verify the truth of such a negotiating position as advanced by diplomats from non-democracies. In short, this argument posits that diplomats representing a dictatorship are less able than representatives from a democracy to claim credibly that domestic pressures preclude them from making certain deals at the negotiating table. For our purpose, this argument also means that, theoretically speaking, it is much more difficult for Chinese and Soviet delegates representing one-party states in boundary talks than Japanese, Taiwanese, Indian and post-Soviet Russian negotiators to resist making concessions by pointing to legislative or popular disapproval.