Constructing A New Global Order:

A Project Framing Document

Background
The existing global political-economic order has been ruptured by a number of recent trends. On
the political side, declining U.S. power (both hard and soft), the return of multipolarity, and,
most importantly, the rise of China as an increasingly important geopolitical player beyond East
Asia has upended a global system previously dominated by the United States and its allies. On
the economic side, financial crises, creeping protectionism, the backlash against globalization,
concerns about surrender of sovereignty to trade agreements or (in the case of Europe) to
regional integration arrangements, and increasing tensions with China on a multitude of trade
and investment fronts has undermined the post-1990 model of hyper-globalization. The “end of
history” consensus that sustained the liberal international order following the fall of the Berlin
Wall has been shattered not only by China’s authoritarian practices and its state-directed
economic model, but by the rise of ethno-nationalist populist movements in the U.S., Western
Europe, and many middle powers. Although Trump’s unilateral actions, contempt for
international organizations, and confrontational approach to China have accelerated these trends,
he is as much a symptom and reflection of the undercurrents destabilizing our present
arrangements as an independent cause.
Covid19 is deepening fault lines in the global landscape even more. The retreat from hyper-
globalization has turned into a leitmotif, as governments talk about bringing supply chains home
and the economic conflict with China deepens. At the same time, the crisis has highlighted how
woefully inadequate our present global arrangements are in providing for global public goods
such as fighting health pandemics or addressing climate change.
How should governments respond to these accumulating tensions? Is it possible to reconstruct a
workable global political-economic order? And if so, what principles should underpin it?
We can imagine different broad narratives on these questions. One approach would be to move
back toward the status quo ante by revitalizing the Western alliance, strengthening multilateral
institutions ranging from NATO to the WTO, and reviving hyper-globalization (albeit with better
distributive outcomes at home). Adherents of this view would have to assume that the
international liberal order can relatively easily coopt the emerging powers into that order without
making significant concessions. A willingness to accommodate China’s geopolitical ambitions to
some extent would be the quid pro quo for preserving the economic benefits of closer integration
through pre-existing multilateral institutions and practices.

1
We are grateful to Henry Farrell, Jeffrey Frankel, Jeff Frieden, Jolyon Howorth, Jeff Lehman, John
Mearsheimer, and John Ruggie for very useful comments on an earlier draft. None of them should be held
responsible for the views expressed here.
1
An alternative narrative would be that the status quo ante is unattainable (due to China’s rise and
other trends noted above) and that the geopolitical and economic threat posed by China requires
a hardline response. Such a response would entail extensive economic decoupling from China, a
sustained military buildup, and an overt strategy of containment. In this vision, the norms or
rules shaping the future world order should be fashioned primarily to give the United States and
its allies significant advantages over China and its own partners. Although this approach is not
intended to lead to war, cooperation with Beijing would be minimal-to-non-existent.
The starting point of our project is the belief that both of these two positions are undesirable.
Regarding the first option, we believe an attempt to return to the status quo ante would be neither
possible nor appealing. A return to hyperglobalization would re-create many of the adverse
conditions that undermined it before 2016, along with the additional vulnerabilities exposed by
the pandemic itself. Equally important, accommodating Chinese ambitions in Asia and in a
number of areas of advanced technology solely for the sake of immediate economic gains
ignores the long-term effects of such a policy on the overall balance of power between
Washington and Beijing. A world in which China enjoyed a dominant position in Asia and a
substantial margin of superiority over the United States would be a world in which U.S. security
is reduced and a substantial price is paid in terms of human rights abuses..
As for the second option, the hardline position not only sacrifices most if not all of the evident
gains from trade and investment relationships with China, it may make military conflict more
likely through the workings of the familiar security dilemma.2 An overly bellicose approach to
China may also alarm the allies whose cooperation we seek and thus leave the United States in a
weaker position. Most important of all, viewing Sino-American relations as a strictly zero-sum
game will make it difficult-to-impossible to cooperate with China on issues where our interests
overlap, such as climate change or global public health.
This project will explore a third possibility, one that seeks the best possible outcomes within a
world order whose outlines will be nonetheless be shaped primarily by Sino-American rivalry. It
proceeds from the recognition that any sustainable global order needs an “operating system”: a
system of rules to help states manage trade, investment, communications, the movement of
peoples, the conduct of diplomacy, and host of other issues. These rules can be embodied in
international organizations or formal agreements or internalized as norms that guide individual
states’ behavior. They will be most effective when they are broadly legitimate; that is, when
they are viewed as desirable and fair by most if not all of the participants.
In practice, the rules that prevail at any given moment are invariably a reflection of the
underlying balance of power and the interests of the major powers. Properly designed, they can
facilitate cooperation and help states achieve gains that would be difficult or impossible to obtain
in their absence. The objective of our project is to develop a realistic and pragmatic approach for
developing rules appropriate for the post-Covid19 world, a world where rivalry between the
United States and China will be a key defining feature. Our ultimate goal is to identify a set of

2
The security dilemma is to the tendency for one state’s efforts to make itself more secure (e.g., by
increasing defense spending) tends to make other states less secure. See John Herz, “Idealist
Internationalism and the Security Dilemma,” World Politics 2, no. 2 (1950); and Charles L. Glaser, “The
Security Dilemma Revisited, World Politics 50, no.1 (October 1997).
2
institutional arrangements that can deliver as much stability and cooperation as possible within
the constraints imposed by the distribution of power and the resulting strategic rivalry.
A ‘Modus Vivendi” Model of World Order
A truly global order will need to acknowledge, and perhaps accommodate, the roles and interests
of many others beyond the U.S. and Chinese governments – not only other states but also a large
number of transnational actors, such as multinational corporations, transnational social
movements, or international NGOs. The presence of these and other non-state actors complicates
the landscape of international relations: it creates a multitude of interests that don’t neatly align
with the borders of nation states; it renders state action potentially less effective; and it also
creates the possibility that many desirable social and economic objectives can be accomplished
through their actions, rather than through government policy.3
In light of these background conditions, an approach such as ours that focuses primarily on the
nation state and inter-state relations needs justification. We eschew a full-throated defense here,
and instead make two quick points that we hope is enough to render what’s to come of interest
even to those who would stress the primacy of non-state actors. First, corporations, banks, and
NGOs are ultimately backstopped by rules enacted and maintained by governments. Put crudely,
if corporations have privileges such as limited liability or legal personhood, it is because states
have given these to them. Financial institutions based in New York City or London would be
unable to operate globally without the financial regulations of the U.S. and U.K.4 Staffs of large
international NGOs must rely on the protections of the states of which they are citizens or in
which they operate. In other words, the thick network of transnational relations that has been
built over decades is not only the product of state actions and regulations; it would not exist
without them.
Second, it is more than likely that the future of international relations will hinge critically on how
the looming conflict between the two strongest powers in the world, the U.S. and China, will be
managed. If we focus here on the U.S.-China intergovernmental relationship, it is because it
presents the test case for the framework we outline below. Any order that does not adequately
address this particular axis has no chance of being a “global” order. If the U.S. and China are
able to reach a modus vivendi in some important areas, on the other hand, we can hope to see our
way around the other dimensions of a global regime.
This is not to say that other states are unimportant. Indeed, as discussed below, there are
important realms where cooperation must include most, if not all, of the states in the system.
3
See Henry Farrell and Abraham L. Newman on the transnational corporate economic connections that
define today’s interdependence (e.g., Farrel and Newman, “Chained to Globalization: Why It’s Too Late
to Decouple,” Foreign Affairs, January/February 2020); and John G. Ruggie on the possibilities of
international corporate social responsibility (e.g., Ruggie, “Corporate Globalization and Its
Consequences: Disembeding and Rembeding Governing Norms,” draft chapter for Peter J. Katzenstein
and Jonathan David Kirshner, eds., Liberalism’s End? Populism, Authoritarianism and the Downfall of
the American Order, Cornell University Press, 2020).
4
As Mervyn King put it in the aftermath of bank rescues during the global financial crisis, “global banks
are international in life but national in death.” See Katharina Pistor, The Code of Capital: How the Law
Creates Wealth and Inequality, Princeton University Press, 2019, on the national legal underpinnings of
international finance.
3
Moreover, countries such as Russia or India and actors like the EU are bound to play important
independent roles in some critical policy areas, and the United States and China will undoubtedly
try to construct thicker cooperative relationships with countries that align with them for reasons
of shared values or other common interests. Even in a world order that is heavily shaped by
competition between the two most powerful states, what other countries choose to do will still
matter.
We begin by recognizing that the U.S. and China are bound to compete for power and influence
in various ways. We recognize further that the United States and China have sharply differing
values and that each has domestic political and economic arrangements (as well as regional
interests) that it will strive to protect. Unlike the hardline vision sketched above, however, our
approach does not preclude cooperating with China in certain critical areas. Nor would it make
the desire to exclude or weaken China the sine qua non of future trade or investment
arrangements, or require other U.S. partners to forego all forms of collaboration with Beijing.
Furthermore, this vision of world order anticipates that both sides would have to adjust their
behavior in certain ways and avoid actions or rhetoric that challenged the other’s domestic
legitimacy.
It follows that a future world order must address two imperatives. First, it will require a set of
norms or mutual understandings between Washington and Beijing, to minimize the risk of
serious conflict while permitting mutual gains in those areas where cooperation is still essential.
Second, a future order also requires rethinking the norms and institutions among those countries
that will be more closely aligned with the United States. Some of these arrangements may be
shaped by the desire to compete more effectively with China, but others should be designed to
address the problems created by higher levels of political and economic interdependence with
these countries and would not be intended to deliberately exclude or penalize Beijing.
Developing more effective regimes on trade, investment, the digital realm, human rights, etc.,
will strengthen ties among those states who are already share many of America’s core values. In
theory, such arrangements could be fashioned in ways that would encourage China to participate,
assuming it was willing to do so.
We expect the emerging global order to be relatively thin, with cooperative arrangements limited
by the increasingly competitive relationship between the United States and China. At the global
level, cooperative arrangements will be confined to particular areas and the depth and
institutionalization of the relevant norms or rules will be modest but not zero. Moreover, in some
areas—such as efforts to field superior military forces or produce certain types of advanced
technology—that rivalry will be mostly if not entirely zero-sum. Within this largely “realist”
global order, the United States and China will each try to construct and manage their own partial
or “bounded” orders based on their own preferred principles, Within the two bounded orders,
like-minded states will apply deeper rules to manage more extensive relations of
interdependence.5
We want to explore whether it is possible to develop a global regime that is stable and relatively
peaceful, respects the diversity of institutional arrangements around the world, reaps the gains of
mutual cooperation among nations (and China in particular) where possible, and remains true to

5
This terminology (and the logic that underpins it) is drawn from John J. Mearsheimer, “Bound to Fail:
The Rise and Fall of the Liberal International Order,” International Security 43, no. 4 (Spring 2019).
4
the values of human rights and democracy to which many governments in the West and
elsewhere have been committed. One might call this a “modus vivendi” model of world order.
We highlight three features such a global regime must display and the questions that each raises
for further analysis.
1. Stability. The new global order must minimize the risk of war and other types of conflict
(including major trade wars). The difficulty here is that the economic and political rise of China
has rendered the existing equilibrium unstable, raising the prospect of what Graham Allison has
called “the Thucydides trap.” Our hypothesis is that avoiding the trap will obligate the U.S. and
the West to recognize and accept both the increased power and influence of China in certain
domains and the fundamentally different nature of its political-economic regime. At the same
time, China will have to recognize that using its greater power to undermine other countries’ core
national and economic interests will inevitably trigger greater global opposition and increase the
risk of violent conflict.
Imagining a potential settlement of this sort raises many questions. What are the new geopolitical
arrangements that China’s increased power calls for? Where are the “red lines” that both sides
would be wise to respect? What are the core national and economic interests of the U.S. and
other Western nations? What are the “rules of the road” that will allow nations to settle their
differences when they disagree, as they necessarily will? We provide a few broad answers to
these critical questions below. Fleshing out the details of a global regime of “live and let live” is
one of the key objectives of the project.
2. Cooperation to reap mutual gains. While China and the West may remain strategic rivals,
there are also many areas of mutual gain. International commerce and investment -- in non-
strategic, non-defense-related areas -- and the provision of global public goods in health and
environment are two such key areas. Decoupling scenarios are pessimistic on the prospects here
because of the evident disagreement between the U.S. and China, as well as among other nations,
on the rules that should govern cooperation in such areas. Trade conflict, driven by the charge
that China does not play the existing rules, is a key example. With considerable justification,
China may decline to play by the rules written by the U.S. and its allies. So even where there are
the possibilities of mutual gains, new and mutually acceptable rules will be needed.
The likelihood of serious Sino-American rivalry should not lead us to overlook the many areas
where global cooperation is badly needed, irrespective of the state of Sino-American relations.
First and most obviously, all nations have a common interest in taming the coronavirus and
developing better means to prevent or contain future pandemics. Second, the United States,
China, and all the other major powers have a common interest in addressing climate change.
Third, the United States, China, Russia, and several medium powers share a common interest in
improving the security of the world’s nuclear weapons and fissionable material. Lastly, human
welfare will also be improved if the international economic order were organized in a way that
encourages economic openness, but in a manner that allows individual nations greater latitude to
control the pace of change and to preserve their own political values and “ways of life.”
Our hypothesis is that it is still possible to preserve or expand the zone of cooperation in these
positive-sum areas by developing a new set of rules that are both more respectful of each
nation’s sovereignty – and hence allow countries greater autonomy in selecting economic and
other policies that may have cross-border spillovers – and draw clearer red lines around “beggar-

5
thy-neighbor” policies or other particularly damaging actions that countries should strive to
avoid. Developing these rules of engagement in different domains – economics, data, public
health, climate change, and some areas of national security – will be another goal of the project.
3. Respect for human rights and other core values. National security and foreign economic
policies cannot be completely separated from nations’ core values. The U.S. government has not
been a consistent defender of human rights abroad and its claims otherwise can be chalked up to
hypocrisy. Nevertheless, human rights abuses and political repression in China are an important
concern aong Western publics, and rightfully so. A stable global order in which nations are able
to reap mutual gains from economic and other interchange need not be indifferent to human
rights. The question is: how and where should these values be reflected? Are there issues or
domains where we should not seek a modus vivendi because doing so runs counter to the values
we hold dear?
We start from the position that this is an unavoidable tension that nonetheless needs to be
managed. How human rights considerations should play into liberal regimes’ dealings with
China is another question that our project seeks to clarify. A critical line to draw may be between
acts of cooperation that directly contribute to maintaining or upholding a repressive regime and
acts of cooperation that do not. The U.S. would avoid the former, but not the latter.6 For
example, the United States might refuse to sell China riot gear or surveillance software, but not
soybeans or commercial aircraft. Needless to say, it is far from immediately clear how such a
distinction would be implemented and maintained across the full range of potential Sino-
American relations. And of course, China and other important global actors may have their own
red lines with regard to practices that are highly valued in Western or other nations (though it is
not yet clear to us what those may be.)
Building a New World Order: A Framework for Analysis
To imagine how a relatively stable world order might be constructed, we propose a meta regime
that presumes only minimal agreement among states (and the U.S. and China in particular) at the
start. This meta regime is essentially a device for structuring a conversation around states’
agreements and disagreements. It is open-ended with regard to actual rules to be applied in
particular domains. So it imposes few constraints on states that want to maintain their freedom of
action; the entry cost is low. But it can nonetheless support significant cooperation. Perhaps more
importantly, it can encourage an increase in cooperation over time among adversaries, as
participation in the meta-regime builds trust between them. This approach allows both thin and
thick versions of international regimes.

6
This is akin to Avishai Margalit’s definition of a “rotten compromise” as “an agreement to establish of
maintain an inhumane regime.” Some arrangements with a human-rights violating regime are truly rotten
in this sense, and therefore should be rejected. But an important implication is that not all compromises
need to be rotten by this definition. Margalit, On Compromise, and Rotten Compromises, Princeton
University Press, 2009.

6
The threshold condition for participation in this meta regime is that states agree to a four-fold
classification of policies – without necessarily agreeing in advance on which actions belong in
each category. The four categories are described below.7
Category 1 refers to principles that all parties accept and actions that they can all agree are
illegitimate or wrong. The latter would be harms imposed on other states that cannot legitimately
be justified by economic or national security considerations of the state that causes them. Such
policies and actions would be prohibited. We shall discuss possible illustrations below.
Category 2 contains policies that do not fit in the previous category – because of insufficient
agreement on whether the action in question is clearly illegitimate – but for which a positive-sum
solution may still be possible. If State A adopts a policy that is harmful to State B, the two
parties may be able to negotiate a mutually beneficial bargain that leaves both better off. The
bargain may involve, for example, State B offering a concession in another domain in return for
State A revoking the policy. We call this the “cooperative negotiations” category.
Category 3 refers to policy domains where cooperation proves impossible, and each state resorts
to its own independent policy action. When two or more states are unable to reach a mutually
beneficial bargain, each is understood to be free to adopt “well-calibrated” national actions to
reduce the harm to its own interests. Importantly, however, such responses should be clearly
linked to the damage being done by the other side’s policies and intended solely to mitigate its
negative effects. In particular, countervailing policy responses should not be undertaken with the
express purpose of punishing the other side or weakening it in the long run. Nor should failure
to reach an acceptable compromise in one area be used as a pretext to retaliate in a different and
unrelated domain. We call this the “autonomous policy” category.
Lastly, Category 4 (the “multilateral governance” category) deals with situations where A’s
bilateral policies toward toward B have significant spillover effects on C, D, or others. Under
these circumstances, an effective response may require involvement and buy-in by most if not all
of the states affected, often under the aegis of a formal multilateral institution. We anticipate a
similar process as outlined under Categories 1-3 to apply, but multilaterally.
The categories in which a particular policy or issue is handled will be influenced by the overall
state of relations between the interested parties, their relative power, and their particular interests.
Between major rivals, for instance, one would expect relatively few issues to land in Category 1
and most to end up in Category 3. Among close allies with similar values, by contrast,
Categories 1 and 2 will be fairly full and Category 3 will contain fewer issues. If we are correct

7
This framework is based on U.S.-China Trade Relations: A Way Forward (2019), available at
https://www.inet.econ.cam.ac.uk/files/us-china_trade_joint_statement_2019.pdf, an initiative of legal
scholars and economists led by Jeffrey S. Lehman, Dani Rodrik, and Yang Yao. Like our current project,
that report sought to define a middle ground between the view that China should be forced to undertake
major domestic reforms in the areas of intellectual property, state intervention in the economy, etc., and
become more like the rest of the OECD, and the view that the United States and China should “decouple”
their intertwined economies almost entirely in order to preserve their preferred domestic orders. The
report laid out a third option, one intended to preserve many of joint benefits of trade while allowing each
side “considerable latitude at home to design a wide variety of industrial policies, technological systems,
and social standards.”

7
that the future world order will be a thin “realist” order at the global level and contain largely
separate and thicker bounded orders organized around the major powers, then the distribution of
issues across the four baskets will vary greatly depending on which states/issues are involved.
We do not expect the boundaries between our categories to be adjudicated by international
institutions or outside enforcers – except perhaps in the case of close allies. The categorization
can nonetheless serve a useful function insofar as it encourages rivals to explain their actions,
clarify their motives, and justify their decisions. If doing so prevents conflicts both sides would
like to avoid, it also serves the longer-term interests of the parties. In other words, a structure
along these lines can “bootstrap” a degree of cooperation that might not have otherwise existed.
Instead of viewing relations between states as one of either “rivalry” or “cooperation,” this
framework encourages us to distinguish between 1) issue areas where there may already be
considerable agreement, 2) issue areas where differences now exist but are at least potentially
resolvable through negotiation and adjustments, and 3) areas where independent national
responses can protect particular national interests without escalating the overall level of conflict
unnecessarily. We do not exclude the persistence, under this scheme, of issues that remain
unreconcilable and that continue to poison inter-state relations. We also recognize that states
will sometimes pursue policies adopted for the express purpose of weakening a rival or gaining
an enduring advantage over it. This feature of international politics does not disappear under our
approach, either for the major powers or for many others. Our hope is that the framework offers
a path for guiding as many contentious issues as possible toward one of our categories, and
hence rendering them less malign for the international order.
Our hope is that viewing relations in this way will facilitate preserving those areas where
cooperation remains necessary and mutually beneficial, even in an era characterized by rising
competition. The United States and China could use this framework to set the their bilateral
diplomatic agenda, and it could also inform cooperative arrangements between the United States
and the countries with which it is likely to be more closely aligned. Moreover, the framework
allows a dynamic evolution of the degree of cooperation between adversaries. One consequence
of a conversation structured along the lines we propose is that it enables parties to build
reputation, develop trust, and better understand the preferences and motives of each other.
Categories 1 and 2 may become more densely populated over time as a result. Finally, the
approach has the additional advantage of being equally applicable to economic issues and
security issues, as we discuss below.
Putting the Framework to Work
Category 1: (The “prohibited” category)
There are some broad principles on which all states could agree from the outset. These might
include, at a minimum, commitments made under the United Nations Charter, the domestic
legitimacy of national governments, and the authority of each nation’s government to regulate
the entry of people, goods, capital, and data into its respective country. This category would also
contain any other policies that are understood by all parties to be illegitimate or wrong.
In the area of economic relations, there are good reasons why states may want to prohibit a type
of policy that economists call “beggar-thy-neighbor” policies. These are defined as “policies that

8
seek to increase domestic economic welfare at the expense of other countries’ welfare.”8 The key
notion here is that the benefit at home comes “at the expense of other countries.” Unlike other
domestic policies that may also entail negative repercussions across the border, “beggar-thy-
neighbor” policies create domestic gains only to the extent that other nations lose.
An example can clarify what is at issue. Consider two different policies, a production subsidy
and an export tax. Both policies can hurt a trade partner: in the first case, because the partner’s
own firms may experience a competitive disadvantage, and in the second case because the
partner’s firms using the taxed product as an input now face higher input costs. However, in
general the production subsidy cannot be considered a beggar-thy-neighbor policy because the
benefits that the home country expects to obtain can be reaped even if its trade partner is not
harmed at all. For example, the objective for the subsidy may be to internalize technological
learning externalities, which can be achieved regardless of impacts on other countries. The
export tax, by contrast, is a beggar-thy-neighbor policy when the home country has global
market power in the product in questions. The benefits that accrue to the home economy in this
instance arise directly from the (terms-of-trade) costs suffered by the other country. There would
be no gains at home without the costs imposed abroad.
The latter is not a hypothetical example: starting in the 2000s, China implemented a range of
export restrictions on rare earth minerals, for which the country is a globally dominant supplier.
The restrictions sharply raised the prices in other markets.9 Other examples of beggar-thy-
neighbor policies include global tax havens (i.e., where corporate tax rates are set artificially low
in order to attract shell companies, at the expense of tax revenue in those corporations’ home
countries) and mercantilist currency policies (i.e., the deliberate undervaluation of the currency
in order to expand employment at home at the expense of unemployment abroad).
A second critical feature of beggar-thy-neighbor policies is that they are negative sum for the
world as a whole. In the export restrictions case, for example, non-competitive conduct creates a
market inefficiency. Except for a few special cases, all countries would be hurt if they were all
free to engage in those policies.10 For example, we could imagine the U.S. retaliating to China’s
export restrictions on rare earths by imposing a similar export tax on agricultural products on
which China relies, leaving both countries worse off. So the general prohibition of such policies
can be motivated on the basis of self-interest of the parties themselves, without having to rely on
global norms.
In the real world, of course, states may legitimately or opportunistically disagree over whether a
particular economic policy neatly fits into the beggar-thy-neighbor category. For example, a
country maintaining very low corporate taxes can argue that the objective is not to act as a tax
haven but to stimulate domestic investment, regardless of source. The U.S. might brand Chinese
industrial policies as beggar-thy-neighbor efforts to seize market share from overseas companies

8
Princeton Encyclopedia of the World Economy.
9
Chinese restrictions were eventually revoked after the WTO ruled that they were inconsistent with trade
rules. See Congressional Research Service, “Trade Dispute with China and Rare Earth Elements,” June
28, 2019.
10
There are some exceptions. For example, when countries are very asymmetric in size, it is possible for
the larger country to be better off in a retaliation equilibrium than in the free trade equilibrium.
9
that are more efficient, whereas China may describe those same policies as developmental
policies intended to stimulate domestic economic growth. China may brand U.S. restrictions on
inward investment by Chinese companies as a beggar-thy-neighbor effort to prevent access to
advanced technologies, while the U.S. may view those same policies as critical for protect
national security. Such disagreements will surely arise. But the framework proposed here
provides a conceptual vocabulary for discussing whether such policies are per se problematic
(i.e., beggar-thy-neighbor). In the case of blatant beggar-thy-neighbor policies, refusal to accept
them as such and leaving them in place would come at some reputational cost, undermining a
state’s negotiating capital and drawing opprobrium from third parties (other states, private
entities such as corporations, and professional legal and economics communities).
Similar principles can be found in the areas of national security and foreign policy. For example,
the UN Charter contains a series of prescriptions and prohibitions that all members of the United
Nations regard as legitimate and have publically committed to follow. For example, Chapter II
of the Charter bars the acquisition of territory by conquest and Chapter VII identifies the
conditions when states may legitimately use armed force (i.e., in self-defense). The United
States, China (and all other UN members) recognize that being judged to have violated this norm
may carry non-trivial political consequences.
Other examples within the “prohibited” category might include direct violations of the principle
of diplomatic immunity (e.g., attacks on embassies, consulates or the unlawful detention of
diplomatic personnel) or an attack on another country’s ships or aircraft on the high seas or in
international airspace. A less formal prohibition (akin to the informal restraint that the United
States and Soviet Union showed during the Cold War) would be the norm that major powers
should strive to avoid direct military engagements with military personnel of other major
powers.11 The actual use of a weapon of mass destruction (and especially a nuclear weapon) fits
in this category as well, despite the lack of formal international convention against it.
Most importantly, there is little or no disagreement between the United States, China, and other
major powers on these principles. Of course, to say that certain actions are proscribed by formal
agreements or informal norms does not mean that violations never occur; it simply means that
the major powers understand where the lines have been drawn and recognize that crossing them
is a serious step. There will inevitably be gray areas where major powers will disagree on
whether an agreed-upon norm applies: for example, China regards Taiwan as an invisible part of
its national territory and might therefore view military action against Taiwan as permissible
under international law, but the United States (and many other countries) would undoubtedly
interpret this event as an illegal act of aggression. Furthermore, a consensus that certain actions
are illegitimate and formally prohibited does not preclude the use of salami tactics, hybrid
warfare, or other measures designed to evade or “design around” an existing norm. Even so,
such evasions are themselves a tacit acknowledgement that the “lines matter” and that open
defiance would entail greater costs.

11
For example, Russian and U.S. forces operating in or near the Syrian civil war have sought to avoid
direct engagement or accidental exchanges with each other.
10
Category 2: (The “cooperative negotiations” category)
This basket contains areas where the United States, China, and some other countries have
conflicting polices and/or interests, but where each has reason to believe it could get the others to
alter their behavior in exchange for other concessions or adjustments.
In economics, there are many policies that might fit in this category. A classic example is import
restrictions that are maintained not to exercise market power on world markets – i.e., are not
beggar-thy-neighbor – but for domestic political reasons such as protecting profits or
employment in a particular domestic industry. A trade partner might offer to remove its own
tariffs in industries of interest to the first state’s export industries in exchange for the elimination
of these restrictions. When both states expect to be better off in the absence of the relevant
import barriers, they can strike a bargain. Or take the case of the subsidy discussed earlier. The
adversely affected trade partner may offer the subsidizing state the following deal: I will relax
some of the foreign investment regulations to enable greater market access for your companies if
you remove the subsidies you are providing to domestic firms. Such exchanges are the basis for
most trade agreements, and the reasoning continues to apply even when states are adversaries in
military or geopolitical realms, unless significant advantages are conferred to one of the parties
in those realms from the maintenance of the trade barriers in question.
Within the security realm, arms control is the classic illustration of this kind of mutual policy
adjustment. Instead of wasting money and/or increasing the risk of war through relentless arms
increases, potential rivals may be able to reach agreements that eliminate specific areas of
vulnerability (thereby making both more secure) and to devote resources that might have been
spent on armaments to other needs. It is easy to think of other examples: great powers could
agree to limit arms sales to third parties (such as warring factions in a civil war), or to cooperate
to make it more difficult for certain states or terrorist organizations to obtain weapons of mass
destruction. As is typical in such cases, the incentive of each party to reap the benefits of
opportunistic behavior in the short run—for example, by selling arms through middle-men—
would have to be balanced against the gains from longer-term cooperation.
At the height of the Cold War, for example, the United States and Soviet Union jointly sponsored
the 1967 Non-Proliferation Treaty, based on their shared interest in limiting the size of the
nuclear “club.” The two states also jointly sponsored UN Security Council cease-fire resolutions
in order to bring the 1967 and 1973 Arab-Israeli wars to a close. Similarly, a series of dangerous
naval confrontations led the two superpowers to negotiate the 1972 Incidents at Sea agreement,
which significantly reduced the danger of mid-ocean collisions between U.S. and Soviet naval
vessels. In theory, one could also imagine the United States, China, or other major powers
jointly agreeing to limit certain military deployments or activities—such as reconnaissance
flights near the opponent’s home territory—in exchange for concessions by the other side.
Category 3: (The “autonomous policy” category)
Because there is no overarching global authority that can protect states from each other,
international politics is sometimes described as a “self-help” system, where each state must
ultimately rely upon its own resources and strategies in order to survive.12 If two or more states

12
This conception is elaborated most fully in Kenneth N. Waltz, Theory of International Politics
(Reading, MA: Addison-Wesley, 1979).
11
cannot reach a mutually beneficial agreement, then each will try to protect its interests through
its own actions (possibly subject to some of the prohibitions contained in Category 1).
In economics, self help will naturally apply in a large number of policy domains where nations
have different preferences and mutually acceptable bargains may not be available. Such
outcomes are already the default option in a wide range of areas that have not been
internationalized even though cross-border spillovers do exist. Take for example highway speed
controls or educational spending.13 Both sets of policies create possible adverse effects on some
trade partners: a speed limit that is set relatively low reduces that country’s demand for oil on
world markets and harms the interests of oil exporting nations; a country that accumulates more
human capital reduces the gains from trade of other countries with which it competes in skill-
intensive products. Yet states consider themselves entitled to full autonomy in both domains.
Same applies to most areas of consumer and product safety, regardless of implications for trade
partners. For example, states have detailed regulations that producers have to meet in order to be
allowed to sell in their market. An automaker or a toy producer abroad has to abide by those
regulations in order to be allowed to sell in that market. To take another example, it is widely
agreed that a country with strict carbon control measures at home could apply a border tax
adjustment (an import tariff) on carbon-intensive imports from countries without such controls.
In other words, the state reserves the right to keep non-abiding foreign producers out of its
national market in order to uphold the regulations it sees fit for its own national circumstances. It
has the right to block international arbitrage through trade and investment flows from
undermining of national standards. This principle could be extended to prevent downard
international arbitrage of corporate taxation, labor standards, financial regulations, or
environmental regulations.
While protecting domestic standards and regulations is an acceptable objective, using domestic
economic policies in a punitive fashion or to deliberately alter the policies of other nations is
much less so. Hence this category legitimizes policy autonomy to the extent that is targeted on
domestic objectives and “well-calibrated.” It is permissible for a state to ban imports of toys with
more than a certain lead content to protect domestic children in line with its own cost-benefit
calculus. It is not permissible to use the import ban as an instrument to alter other countries’
lead-content regulations or to use them to escalate a trade war. In practice, it may be difficult to
disentangle the two cases; but there is an important difference in principle. It should not be
difficult to conclude, for example, that Trump’s trade was with China has violated this principle
– by Trump’s own statements of intent.
Major powers typically rely largely on their own national efforts in matters of national security
as well, although they may cooperate with and fight alongside close allies when necessary.14 If
arms control efforts fail, then rival states will strive to enhance their own defense capabilities and
conduct R&D designed to gain or preserve a technological edge. Rivals will engage in various

These and other examples are discussed in Dani Rodrik, “Putting Global Governance in its Place,”
13

World Bank Research Observer, vol. 35(1), February 2020.

12
forms of espionage to determine what others are planning, to steal military secrets, or to gain
other advantages. Rivals will also try to increase their influence with other key countries, form
alliances with some of them, or undermine the cohesion of an opposing coalition. As long as
there is no central authority to protect states from each other, great power competition is not
going to stop.
In theory, these and other competitive activities could be addressed in the Category 2
(“cooperative negotiations”), and some of them will undoubtedly arise at summit meetings or in
the normal conduct of great power diplomacy. But history suggests that meaningful agreements
on many of these issues will be difficult to achieve or sustain.15 Arms control agreements
between the United States and Soviet Union involved long and contentious negotiations, for
example, and while they eventually reduced both sides’ arsenals significantly, they did not stop
either side from developing weapons it really wanted, did not eliminate the ability of each side to
completely destroy the other, and did not end their costly efforts to develop the ability to
eliminate the other side’s forces in a first-strike.16 Moreover, some of these agreements (such as
the 1972 Anti-Ballistic Missile Treaty) were subsequently abandoned.
As in the economic domain, in most cases it would be desirable if national responses to a foreign
security challenge were “well-calibrated”: the chosen remedies should be proportional to the
harm inflicted by another state’s potentially threatening actions (including its own efforts to
increase its military power, recruit allies, etc.). This recommendation is designed to guard
against the danger of a tit-for-tat spiral of escalatory steps that goes beyond whatever measures
are necessary to allow interdependent economies to preserve “considerable latitude” over their
own domestic priorities and “ways of life.”
However desirable this principle might be in the abstract, operationalizing it in national security
issues will be difficult, because what one country judges to be a prudent and “well-calibrated”
response could easily look like a dangerous and unnecessary provocation to its opponent. To
make matters worse, each side is likely to prefer a margin of superiority in its favor to an even
balance of power and will therefore be tempted to overreact when faced with some new and
worrisome action by the other. Weapons manufacturers, the uniformed military, hawkish
analysts, and rival politicians will work overtime to exaggerate threats and convince top officials
to do more, making the crafting of a “well-calibrated” response even harder.
Even so, there are prudential reasons why both sides might voluntarily choose to limit their
responses in certain ways. Even in the absence of a formal or informal agreement that placed
certain acts into the “Prohibited” category, great power rivals would be wise to eschew policies
that threaten (or appear to threaten) the other side’s political stability, territorial integrity, or

15
The early 19th century Concert of Europe is often invoked as an example of a cooperative regime that
sought to minimize great power competition, but the formal operations of the Concert broke down in less
than a decade. Historians and political scientists disagree about the extent to which the informal norms of
the Concert helped reduce great power conflict in Europe during the rest of the century.
16
As former Secretary of Defense Harold Brown observed in 1983, "the achievements of arms
negotiations to date have been modest indeed, as are their immediate prospects. . . . In all, not much to
show for thirty-five years of negotiations and twenty years of treaties." Harold Brown, Thinking about
National Security: Defense and Foreign Policy in a Dangerous World (Boulder, CO: Westview Press,
1983), p. 185.
13
national survival, or that suggest its long-term objective is the complete elimination of the other
side as a political rival. Such actions will inevitably provoke worst-case fears on the other side,
and shrink the number of areas or issues where it would be willing to compromise or cooperate.
The bad news is that this objective will also be difficult to achieve, especially given the
significant differences in U.S. and Chinese political values. Although existing norms of
sovereignty and non-interference preclude direct efforts by either power to influence the other’s
internal politics, fundamental differences over core political values will make it difficult to
maintain the “modus vivendi” approach we outlined above.
In particular, differences in values can generate conflict and suspicion even when governments
act with restraint. Alternative political systems can pose a threat merely by existing, because
their very presence constitutes a different model that might inspire reformers or rebels inside the
rival society. This problem will be especially challenging when a political order is based on
universalist principles (such as the claim that all human beings possess certain inalienable
rights), as such claims by their nature transcend political borders. Even if the U.S. government
never lifted a finger to promote those ideals abroad, America’s presence as an embodiment of
these ideals (however imperfectly realized) might be seen as threatening by governments based
on different values.
Furthermore, it is hard to imagine that political communities in either the United States or China
(or many other countries) could consistently stay “on message” and refrain from actions or
statements that suggest an active desire to alter the other’s domestic political arrangements. No
matter what a president does or says, at least a few U.S. politicians can be counted on to
condemn China’s domestic economic practices or its human rights records, and to call explicitly
for it to change. Even if top officials in a particular administration take great care to avoid
provocative statements (and there is no guarantee that they will), there are bound to be prominent
voices on both left and right advocating continued efforts to shape China’s internal evolution.
Similarly, Chinese debates on foreign policy are not monolithic, and American observers looking
for ambitious declarations of China’s desire for global supremacy or harsh Chinese criticisms of
the U.S. system do not have to search very long.17 A “live and let live” approach to each other’s
domestic systems may make good strategic and economic sense, but implementing it consistently
over time will be a major challenge.
The good news, however, is that several other factors may help keep the Sino-American rivalry
within bounds. One obvious sources of restraint is nuclear weapons, which discourage direct
attacks on an opponent and give rival powers ample incentive to tread carefully when crises
erupt. A second factor is a combination of geography and population: the United States and
China are countries of considerable geographic size with large populations and neither could
have any hope of conquering the other. Distance poses a further barrier: although the United
States has military forces deployed all over the world and China’s power projection capabilities
are improving, the ability to attack another great power declines sharply with distance and

17 See, for example, State Information Office, People’s Republic of China, “The Record of Human Rights
Violations in the United States in 2019” (March 2020), available at
https://www.globaltimes.cn/content/1182529.shtml

14
especially if one has to travel across oceans and conduct an amphibious invasion. Although each
could probably contrive some sort of conventional attack upon the other’s home territory (at
present, America’s ability to attack the Chinese homeland is much greater than China’s ability to
threaten the continental United States), neither country could project sufficient military power
across the Pacific to threaten the other’s territorial integrity or independence. In this case, what
John Mearsheimer calls the “stopping power of water” makes each side more secure and should
lower the intensity of their competition somewhat.
These features do not make a clash of arms impossible—especially at sea or over Taiwan—but
they do place limits on what either state could aspire to do or has reason to fear. Proxy wars, a
serious competition for allies and influence, and other forms of geopolitical hardball are likely,
but present strategic conditions discourage an all-out effort against each other. Nor are these
background conditions likely to change significantly in the forseeable future.
Finally, we repeat that our framework has the potential to generate increased trust between the
U.S. and China over time. It is not predicated on the two sides trusting each other at the outset,
but it leaves open the possibility – and indeed increases the likelihood – that they might trust
each other more as they work out other disagreements within its confines.
Category 4 (the “multilateral governance” category)
This category contains policies or issues where the relationship between two states has important
implications for third parties, creating clear incentives for multilateral solutions.
Global public goods are the most obvious case that fits this category, and climate change is the
archetypal illustration. A country’s carbon control policies provide benefits to all other nations,
and none can be excluded from those benefits. Conversely, a country’s emissions harm all other
nations equally. In the absence of multilateral restraints, the incentive for each state is to try to
free ride on the control carbon policies of the others. The result is excessive emissions and rapid
climate change. A second example might be global public health, as COVID19 has reminded us.
The rapid sharing of information about potential pandemic and the development of therapeutic
medicines or vaccines represent global public goods.
In economics, there are few domains that might be strictly considered global public goods, even
though journalistic narratives often suggest otherwise. We are told: “sound prudential regulation
in financial centers is a global public good,” or “an open world economy is a global public
good,” but such nostrums often misuse the term. When the U.S. manages its macroeconomy and
financial system well, or it keeps its markets open, economic logic suggests that U.S. citizens
will be the primary beneficiaries. Other countries may benefit too, but this does not reduce the
incentive of U.S. policy makers to pursue those sound policies. By the same token, when those
policies are mismanaged, as they were prior to the global financial crisis or under President
Trump, it was the U.S. economy that incurred some of the highest costs. Such mistakes do occur
of course. But the argument that international cooperation or multilateral rules can reliably
prevent these mistakes from happening is difficult to make.18 Of course, well-designed
international norms make it harder (though not impossible) for states to do stupid things that
harm themselves (along with others); what is not clear is how states can be expected to develop

18
Rodrik (2020), op. cit.

15
“well-designed” norms internationally when they are liable to make mistakes in those same areas
domestically.19
In the areas of national security and foreign policy, one can imagine two different types of
multilateral arrangements.
One type of multilateral arrangement would include both the United States and China, as well as
other powers such as Russia, India, or the EU. These institutions would facilitate cooperation on
areas where state interests substantially overlap, such as climate change, global public health
(e.g., responses to COVID19 and other future pathogens), measures to enhance secure control
over nuclear weapons and materials, or counter-terrorism. Efforts to strengthen the Law of the
Sea regime or to develop norms to regulate activities in cyberspace might fit here as well.
A second type of multilateral arrangement consists of institutions created, led, or supported by
one of the two main powers but where the other is excluded. Such arrangements would typically
be intended to help the member states deal with problems created by the policies of states that
are not part of the agreement. Formal military alliances are an obvious example of this type of
institution, but as previously discussed, both the United States and China are likely to try to
create like-minded coalitions of countries that coordinate their economic and security relations,
adopt similar norms regarding digital privacy and surveillance, and use similar digital
technologies. The abortive Trans-Pacific Partnership would have been one such arrangement,
and its strategic goal (i.e., to cement closer ties with a number of Asian countries) may have been
more important than any of its economic provisions or impact. There is a growing possibility
that the digital world will be increasingly divided into Sino-centric and Western-centric realms—
what some analysts have dubbed the “splinternet”—and this would be another case where
multilateral cooperation expanded but on a partial rather than global basis.20
An Illustration: The Huawei Case
To illustrate how our approach might help address a concrete, real-world case, we offer here a
brief discussion of the current dispute over Huawei 5G technology. The question is: can our
framework suggest how to deal with concerns in the West that Huawei poses a national security
threat and the measures that the U.S., in particular, has taken in response?
Huawei is a company that is nominally owned by its workers, but has long been suspected by
Western analysts to have close ties to the Chinese security establishment. Starting from humble
beginnings and complete reliance on reverse-engineered Western technology, the company has
become one of the world’s largest telecom vendors and a key player in the development and
installation of next generation (5G) networks. The U.S. has sought to cripple the company’s
international activities since the early 2000s, when Huawei first tried to enter the U.S. market. It

19
One generic exception to this skepticism is provided by dynamic inconsistency, which arises when
actors understand they will be tempted to make mistakes in the future and have the incentive to take an
action at present to reduce that likelihood. Signing on to international agreements may then allow states to
make (sort of) binding commitments that effectively prevent their current selves from harming their future
selves.
20
“The ‘Splinternet of Things’ Threatens 5G’s Potential,” The Economist, December 25, 2019, at
https://www.economist.com/the-world-in/2019/12/25/the-splinternet-of-things-threatens-5gs-potential

16
prevented Huawei from acquiring American companies through the CFIUS (Committee on
Foreign Investments in the United States) national-security review process. It pressured
American telecom operators not to work with Huawei. The U.S. Congress undertook an
investigation of the firm. Sanctions were imposed on the firm in several rounds, and the daughter
of Huawei’s founder and the company’s CFO was eventually placed under house arrest in
Canada due to legal charges against Huawei in the U.S.21
For our purposes, it is useful to distinguish between two kinds of U.S. actions in particular: first,
the restrictions the U.S. has placed on Huawei’s entry and operation in the U.S.; and second, the
restrictions the U.S. has imposed on American companies to prevent them from supplying chips
and other components that feed directly or indirectly into Huawei’s supply chain. We do not aim
to provide definitive “solutions,” but merely to show how our framework could help manage
what might otherwise seem like an irresolvable and easy-to-escalate conflict.
Consider first the ban on on Huawei’s American operations. In our framework, the justifiability
of this policy rests on the plausibility of the national-security argument. If the argument is
plausible on the face of it, the restraints of Category 1 (“prohibited actions”) would not apply and
the U.S. would be justified in preventing Huawei from operating in the U.S. Our reading of the
evidence is that the national-security argument is quite strong in this case: although there is no
evidence that Huawei has engaged in spying or cyber-security violations, there is enough
uncertainty around its technical capabilities (given the opacity of the software) and the
company’s links to the Chinese government for the U.S. government to take a precautionary
stance.
Having decided that Category 1 does not apply, we might then ask if the case fits under Category
2 (“cooperative negotiations”)? The Trump administration would clearly answer in the negative:
China is too big a threat and too unreliable a partner to engage in give-and-take over sensitive
issues of American national security. But future administrations and other nations may answer
differently. Here the U.K. presents an interesting example of how Category 2 might work in
practice. The British government has entered an arrangement with Huawei under which the
company’s products in the U.K. telecoms market undergo an annual security evaluation. The
evaluations are undertaken by the Huawei Cyber Security Evaluation Centre (HCSEC), a facility
that opened in 2010 and is governed by a board that includes a Huawei representative along with
senior officials from the British government and the UK telecom sector. HCSEC’s reports are
public. The latest concluded that there was enough cause for concern: Huawei’s “approach to
software development” raises risks to UK operators and “requires ongoing management and
mitigation.” It noted: “The Oversight Board advises that it will be difficult to appropriately risk-
manage future products in the context of UK deployments, until the underlying defects in
Huawei’s software engineering and cyber security processes are remediated.”22 In July 2020,

21
“U.S-China: Is Huawei ‘Too Big To Fail?’,” Financial Times, August 21, 2020.
22
HCSEC Annual Report, March 2020,
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/790270
/HCSEC_OversightBoardReport-2019.pdf.

17
Britain decided to ban Huawei from its 5G network – though this seems to have been due
primarily to pressure from the Trump administration and not the result of HCSEC’s work.

Beyond the cooperative element – at least in principle – the British approach is also noteworthy
in that it has a degree of transparency built into it. Since HCSEC’s reports are public, the
technical reasoning on which a national-security determination is made can be seen and
evaluated by all parties. This includes domestic firms who may have a commercial stake in
Huawei’s investments as well as the Chinese government and Huawei themselves. This can help
build mutual trust as the parties develop a fuller understanding of each others’ motives and
actions. The Chinese side may come to appreciate the legitimate concerns that the home
government has. Conversely, it may become more difficult for the home government to feign
national security concerns when the grounds are weak or are a cover for, say, purely protectionist
commercial considerations. By contrast, CFIUS, which has a much wider remit, publishes only a
single annual report that presents statistical information about the cases it has investigated,
without any of the evidence or reasoning behind its judgments.

This discussion suggests that some sort of adjustment between Huawei and the UK government
was theoretically possible, and a full working out along Category 2 principles might have
ultimately led to a settlement or a quid pro quo between the two governments. Even in the British
case, however, this never quite seemed in the cards. Therefore, the most likely scenario seems to
be that the U.S. and other Western governments will choose to make unilateral decisions on
Huawei. This moves us into Category 3, the realm of unilateral but “well-calibrated” responses.

At first glance, the Trump administration appears to have chosen this route, but a closer look
suggests otherwise. While governments are fully entitled to take protective measures under
Category 3, their responses are supposed to be proportionate and to avoid the escalation of the
conflict. The U.S. government has violated these strictures by imposing ever-increasing
sanctions and barriers on the international operations of Huawei. Most critically, it has banned
American corporations from selling chips and other components to Huawei and its suppliers,
regardless of where they operate. The clear intention seems to be to deliver a fatal blow to the
company by starving it of essential inputs.

The export ban on U.S. companies is far harder to justify on national security grounds than the
ban on Huawei’s U.S.-based operations. While we would not dismiss outright the possibility that
such a justification exists, it is far from clear what it would be. If there is a national-security
justification, it would have to be made explicit. Huawei’s operations in third countries may pose
a security risk to those countries; but it is the governments of those countries that are in the best
position to evaluate the risks and consequences of shutting off Huawei’s operations. It could be
that there are spillover risks to the U.S. from Huawei’s presence in countries that are U.S. allies.
Even then, the appropriate response would have been to convince these allies to engage in
collective and coordinated action instead of undertaking unilateral action. It looks to us as if U.S.
action to cripple Huawei in third markets is a classic beggar-thy-neighbor action, which is a path
our framework explicitly forecloses. We conclude, provisionally, that the U.S. has over-reached
and moved outside the boundaries of Category 3 when it comes to the export ban.

18
Moreover, the U.S. ban has serious economic repercussions for other countries. If Huawei were
to fold, the effects would be crippling for national telecoms companies like BT, Deutsche
Telekom and Swisscom and others in no fewer than 170 countries that rely on the Chinese
company’s kits and hardware.23 Leaving lock-in effects aside, poorer nations are
overwhelmingly dependent on cheaper Huawei equipment. According to Liberia’s former
minister of public works, “in a world where Africa’s choices are limited to European and
American telecommunications providers, it is inconceivable that the connectivity and cell phone
penetration we observe today would be possible.”24

These implications suggest that the U.S. export ban also violates the tenets of Category 4, which
pertains to spillovers for third parties. Even if the ban were justified on U.S. national security
grounds, the U.S. might have at least engaged in a multilateral process that recognized the
economic costs to other nations of cracking down on Huawei in third markets. By proceeding
unilaterally in a domain with clear global implications, it has acted outside the boundaries of
Category 4 as well.

In short, our framework would give the U.S. considerable leeway in applying restrictions on
Huawei (or other foreign firms) where operations in the U.S. are concerned. For example, the
ban on the domestic operations of Huawei might even be broadened to other foreign firms, to the
extent that those firms are integrated with Huawei’s supply chains and might pose similar
security risks. However, the framework is less permissive with respect to the export ban on U.S.
corporations and internationalizing the ban outside a multilateral framework.

We might pause to consider whether our framework would make any difference in practice. The
ultimate test is whether it would serve to legitimize justified actions while restraining
unwarranted actions that are harmful to other nations. In international relations, powerful nations
are generally free to act as they want. Clearly, any U.S. (or Chinese) administration willing to go
it alone will do so. The most we can hope is the development of shared norms that help
operationalize the concept of legitimacy and act as informal restraints over time. Smaller states,
non-governmental actors, or academics like us cannot encourage good behavior and discourage
bad behavior without knowing what they would be. So we have to start by sketching what the
relevant norms might be in the first place. Doing so gives us a yardstick for evaluating states’
behavior and a counterfactual for what would have been desirable actions. Furthermore, if those
norms can clarify how it is possible for states to pursue critical national security and economic
objectives without harming other nations’ core interests, they could eventually be perceived by
other political leaders as well as public audiences as broadly appealing. Then there might be
clearer benefits for acting in accordance with the norms, and greater costs for violating them,
creating in turn incentives for norm-reinforcing behavior even by great powers.

23
Financial Times, op. cit.
24
W. Gyude Moore in Quartz Africa https://qz.com/africa/1629078/africa-will-stay-loyal-to-chinas-
huawei-regardless-of-trump/.

19
 Implications
The “Four Categories” framework seems readily applicable to both economic and national
security issues. It also suggests several ways that policymakers hoping to manage conflict and
encourage cooperation between the United States and China might proceed.
With regard to Category 1, the primary goal should be to articulate clearly the agreed-upon set of
“prohibited” actions, clarify ambiguities about them, and to expand the list where it seems
possible. The greater the number of areas of potential conflict that can be taken “off the table”
without harming either country, the better Sino-American relations will be.25 Obvious
possibilities include formal or informal agreements not to support anti-government movements
in either country, or norms against the use of cyber weapons against vital national infrastructure
(power systems, dams, health care facilities, etc.). Formal or informal agreements not to openly
challenge the legitimacy or moral integrity of either country’s political order might be negotiated
as well.26 Even if unsuccessful, well-intentioned efforts to negotiate such understandings would
help identify areas where no consensus exists and clarify where the opponent’s “red lines” are.
To maximize prospects for “cooperative negotiation” in Category 2, U.S. and Chinese officials
could begin by identifying policies adopted by the other side that it regards as particularly
worrisome, and where it is at least conceivable that adjustments could be made. For example,
China is not going to abandon its long-term claim to sovereignty over Taiwan, but it is at least
conceivable that they would consider reducing the size of the military forces poised at the island,
depending on what they might gain in return. Similarly, China will not forego its state-directed
economic model. Greater clarity about those practices that present the greatest costs for the U.S.
economy (and not simply for those corporations with the greatest lobbying clout in Washington,
D.C.) might help identify bargains that seem unavailable when the U.S. is challenging the very
foundations of the Chinese economic model. At the same time, each side should also consider
which of its current activities it might be willing to modify or forego in the event that other side
was willing to compromise on an issue of paramount concern. For example, would the U.S.
military be willing to reduce its intrusive reconnaissance efforts close to Chinese air space and
territorial waters, in exchange for a reduction in Chinese land reclamation efforts and territorial
claims in the South China or East China Sea? More controversially, could the two states reach a

25
This principle applies to other bilateral relations as well. For example, the 2015 Joint and
Comprehensive Plan of Action with Iran sought to remove the question of Iran’s nuclear program from
the active diplomatic agenda (in exchange for lifting some international sanctions), in the hope that Iran
and the other signatories could then address other areas of policy disagreement. Regrettably, the Trump
administration did not allow this effort to continue.
26
It goes without saying that Secretary of State Pompeo’s recent condemnations of the Chinese
Communist Party are inconsistent with such an approach. See “Communist China and the Free World’s
Future,” Speech at the Nixon Library, July 23, 2020, at https://www.state.gov/communist-china-and-the-
free-worlds-future/

20
“grand bargain” involving U.S. concessions on Taiwan in exchange for formal Chinese
acceptance of a long-term U.S. security presence elsewhere in Asia?27
With respect to national policy responses in Category 3, both sides should think harder about
what constitutes “well-calibrated” responses; i.e., actions that are sufficient to protect a vital
interest but are less likely to provoke an escalatory response by the other side. Trade wars a la
Trump would obviously fall outside what is permissible here, designed as they are to punish the
Chinese economy and largely ineffective (or even harmful) with respect to domestic
employment/competitiveness objectives. But in most other areas this is a challenging problem
given the familiar workings of the ‘security dilemma’ (i.e., efforts by one side to make itself
more secure tend to make the other side less secure), and the obvious connection between
Categories 2 and 3 make this even more difficult. In some cases, the threat of unilateral action in
Category 3 is precisely what may persuade the other side to make concessions in Category 2, but
that same threat can also poison relations and invite retaliation instead. As a general rule,
however, unilateral responses in Category 3 should be narrowly focused on the specific issue in
dispute, to avoid signaling unremitting hostility, the desire to gain a permanent advantage, or
worst of all, a commitment to achieving the total defeat of the rival social system.
A further implication is that the United States and China should go to some lengths to keep
channels of communication open, so that top officials have regular opportunities to explain what
they are doing and why, and to explore opportunities for Category 1 or Category 2 agreements.
Ideally, the two states would participate in a regular set of bilateral “strategic dialogues”
involving senior officials, together with less frequent meetings of the respective heads of state.
This practice was begun under President George W. Bush and continued by Barack Obama; it
has languished under President Trump.28 Recreating regular forums of this sort and insulating
their regular occurrence from domestic pressures or the normal vicissitudes of great power
relations would be highly desirable.
Fashioning successful multilateral responses (Category 4) will require close attention to the
interests of third parties. Even states with close security ties to the United States do not want to
get caught in a crossfire between the two major global powers; their overriding objective is to
preserve a tranquil global and regional order and continue to reap the economic benefits of
globalization.29 Third parties are therefore likely to gravitate away from the great power they
deem most responsible for “disturbing the peace,” and to move toward whoever seems most
committed to preserving the territorial, political, and economic status quo. If U.S. efforts to
contain China are too aggressive, reckless, or ill-conceived, its regional partners will question

27
For a proposal along these lines, see Charles L. Glaser, “A U.S.-China Grand Bargain?: The Hard
Choice between Military Competition and Accommodation,” International Security 39, no. 4 (Spring
2015).
28
In their first meeting in April 2017, Trump and Chinese President Xi Jinping agreed to establish four
high-level dialogues to manage the relationship. All four met in 2017; only the Diplomatic and Security
Dialogue met in 2018; none met in 2019. See U.S.-China Relations (Washington, DC: Congressional
Research Service R45898, August 2019).

See the recent article by Prime Minister Lee Hsien-Loong of Singapore: “The Endangered Asian
29

Century: China, America, and the Perils of Confrontation,” Foreign Affairs, May/June 2020.
21
whether a close alignment with Washington is worth the risks. By contrast, if these states see
China as the primary threat to the regional status quo and the country whose actions are raising
the temperature, they will distance themselves from Beijing and be even more interested in
cooperating with the United States. In Category 4, it pays to be nice.
The more progress the U.S. and China can make in Categories 1 and 2 (and the more restrained
their responses in Category 3) the easier it will be for them to cooperate with each other and with
third parties on matters where their interests align. An all-out U.S.-China Cold War will make
agreements on climate, health, terrorism, nuclear security, and other global public goods harder
to reach, because each side will be wary of making a bad deal that gave the other an advantage
and domestic opposition to “dealing with the enemy” will be fiercer. Ironically, the more that
the two states can successfully manage the policy domains where their interests are at odds, the
easier it will be to cooperate on issues where their interests align.
Finally, our framework could also guide efforts to reform or replace existing international
institutions. A new world order will not emerge from a completely blank slate; it will be built on
the foundation of institutions that already exist(e.g., the United Nations, World Bank, World
Trade Organization, etc.), as well as newer institutions such as the Asian Infrastructure
Investment Bank, or Paris Climate Accord). By encouraging states to handle policy disputes
within particular categories, and by enshrining the idea of “well-calibrated” responses, the
“meta-regime” we propose could reform efforts more legitimate in the eyes of many states. If
embraced by the United States, it could help convince other countries that the United States was
genuinely committed to preserving as much mutually beneficial cooperation as possible and to
reducing the risks of dangerous conflicts, while at the same time recognizing that national
governments retain the latitude they need to preserve vital national interests.

22