ECONOMIC INTEGRATION AND GLOBAL GOVERNANCE: WHY SO LITTLE SUPRANATIONALISM?

ECONOMIC INTEGRATION AND GLOBAL GOVERNANCE: WHY SO LITTLE
SUPRANATIONALISM?

Miles Kahler
David A. Lake
University of California, San Diego
mkahler@ucsd.edu
dlake@ucsd.edu


Version 4.0



Contribution to Walter Mattli and Ngaire Woods, editors
Explaining Regulatory Change in the Global Economy

---

ECONOMIC INTEGRATION AND GLOBAL GOVERNANCE: WHY SO LITTLE
SUPRANATIONALISM?

Miles Kahler
David A. Lake
University of California, San Diego
Draft 4.0



Economic models of international governance predict that greater levels of global
economic integration are likely to produce changes on two institutional dimensions. The
site of governance should shift from the national to the regional and global levels as
states pool their decision-making powers. At the same time, delegation to supranational
institutions should be favored for managing the governance requirements of a more
integrated economic order. Not only should states coordinate their responses to global
challenges, the gains from specialized governance and the need for enhanced credibility
and dispute resolution should also lead states to transfer authority to supranational
agents.

Yet, the record of international economic governance over the past quarter-
century of globalization does not support these predictions. Outside the European Union,
the functions and delegated authority of global and regional economic institutions have
not uniformly increased. Regulatory policies in particular have not generally migrated
from national governments to new or expanded supranational institutions. Given the
rapid increase in global economic integration -- whether measured by policy
liberalization, by trade, capital, and migration flows, or by price convergence -- why is
there so little supranationalism?

After describing the puzzling (for economic models) variation in supranationalism
across issue-areas, we present a two-step explanation for this “missing”

---

2
supranationalism. We argue that, in addition to supranationalism, at least two other
modes of international governance exist: hierarchy, in which states transfer regulatory
authority to dominant states for certain limited purposes, and networks, in which states,
private actors, or both share regulatory authority through coordinated and repeated
interaction. Hierarchies and networks serve as functional substitutes for supranational
delegation to international institutions.

In a second stage of our explanation, political models of international governance
are deployed as substitutes for or supplements to economic models of governance.
Economic models predict too much supranationalism and cannot explain variation
among these modes of governance. Efficiency is often overwhelmed as a driver in the
presence of distributional and institutional conflict, which often characterizes regulatory
policies. Drawing on earlier work, political explanations for the apparent “missing”
supranationalism are provided, and preliminary conjectures on the politics of choice
among alternative governance models are presented.1 Political models and the
conjectures that they inspire also offer explanations for the relative frequency of
regulatory capture among these alternative forms of international governance. The
likelihood of regulatory capture is determined by both international and national
variables. None of the governance modes—supranationalism, hierarchy, or networks—
is inherently immune from capture
Globalization and Supranationalism: Unexplained Variation

In contrast to the first era of globalization (before 1914) and the interwar decades
of economic turbulence and closure, planning for international economic governance
after World War II awarded a far more central role to intergovernmental organizations
(IGOs) than had been the case under the League of Nations. The turn toward formal
IGOs, however, was not a result of increased economic integration: those institutions,

1 See Kahler and Lake 2003a, 2003b.

---

3
whatever their liberalizing goals, were designed during World War II and its aftermath, a
low point in international economic integration. As global economic integration
increased, particularly after 1980, its effects on these institutions of supranational
delegation was far from uniform. One favored prediction had been that such integration,
if it did not induce a nationalist backlash, would require substantial increases in the
authority of supranational institutions. Although economic integration proceeded at a
different pace according to region and issue-area, its advance in key areas of trade,
finance, and foreign direct investment could not be questioned. Yet institutional
outcomes at the global level were hardly uniform. Globalization produced a marked
increase in supranational delegation in the trade regime, a decline in supranationalism in
the monetary and financial regimes, and a complete failure to delegate in the rules
governing foreign direct investment.

The trade regime has most closely followed the predicted path of increasing
authority delegated to global and regional IGOs. The scope of the World Trade
Organization (WTO), founded in 1995, was larger than its predecessor, the GATT,
encompassing trade in services, as well as behind-the-border policies that had
previously been labeled “domestic,” or at least outside the scrutiny of intergovernmental
institutions. In particular, regulatory domains, such as intellectual property, health and
safety regulations, and technical standards were all brought under the scrutiny of the
new organization. For the first time, regulatory policies were brought within the trade
regime in their own right and not ”as an adjunct to a larger concern about border trade
issues.”2 Competition policy, rules on foreign investment, and labor and environmental
standards were promoted as new (and traditionally domestic) spheres for future
negotiation; they were already included in regional trade agreements. In each of these
cases, new domestic actors were mobilized, distinct from the old exporter and import-

2 Barton et al. 2006, 209.

---

4
competing constituencies, and the trade regime integrated new principles and norms to
govern these issues.3 The new authority of the WTO was reflected in a more efficient
Dispute Settlement Understanding that reduced the power of national governments to
deflect and delay panel reports and in an Appellate Body that could engage in judicial
lawmaking.4

In the oversight of national financial, monetary, and exchange rate policies—
where global economic integration has progressed farther than it has in trade in goods
and services—supranationalism has retreated since 1945. The International Monetary
Fund (IMF), created with substantial (at least on paper) powers for the oversight of
exchanges rates and exchange restrictions on current account transactions, lost much of
that formal authority after the breakdown of the fixed parity exchange rate regime in
1971-73 and the Second Amendment to the Articles of Agreement of the IMF that
confirmed the new regime. The IMF has been peripheral to exercises in macroeconomic
policy coordination among the G7. Although the IMF and the World Bank have assumed
a greater role in surveillance of national financial regulation, particularly among emerging
market economies, the Bretton Woods organizations are part of a network of regulatory
oversight agencies, and they have not been delegated any substantial new powers in
the regulatory domain. Recent debate over the IMF’s medium-term strategy has
revealed support for some increases in delegation to the IMF, particularly in the realm of
surveillance.5 Despite pressure from the United States, members appear to have little
enthusiasm for a restoration of an independent IMF role in the oversight of exchange
rate adjustment among the major economies.

3 Barton et al. 2006, 94-98, 125-151.
4 Ibid., 75-87.
5 The International Monetary and Financial Committee (IMFC) has emphasized greater
independence in the surveillance function, greater transparency, and a larger reliance on the
Independent Evaluation Office. (IMF Survey, 1 May 2006, 118-119).

---

5

If supranational delegation has increased in the trade regime and declined in the
monetary and financial domain, the international rules governing foreign direct
investment, a major driver of globalization, reflect yet another distinct pattern. The old
regime of investment protection, maintained through imperial rule and then through
sanctions administered by the major powers, began to crumble in the 1970s. In
response, the United States and other industrialized countries began to negotiate
bilateral investment treaties (BITs), some 153 in the decade from 1977 to 1986. A
second inflection upward in BITs occurred in the 1990s, as former socialist economies
used these agreements to signal their embrace of capitalism.6

This web of bilateral treaties, now totaling more than seventeen hundred, was a
less efficient means to the end of investment protection and liberalization than a
multilateral investment regime. Negotiations to produce a Multilateral Agreement on
Investment (MAI) at the OECD failed in 1998, however, in the face of a concerted
campaign by NGOs and weak support from the corporate sector. The Doha round of
trade negotiations at the WTO included the relationship between trade and investment
as part of its original agenda, but its was dropped in the July 2004 package agreement.
The OECD Code of Liberalization of Capital Movements remains the “only multilateral
framework in force on international capital flows, including FDI.”7 The WTO oversees
several investment-related codes, in particular, the agreement on Trade-Related
Investment Measures (TRIMS) and the General Agreement on Trade in Services
(GATS). The degree to which regulatory authority over foreign direct investment has
migrated to the global level and has been delegated to IGOs is minimal, however. This
leaves the network of BITs as the primary governance structure in this issue-area.
Alternative Modes of Global Governance

6 Vandevelde 1998, 628.
7 Golub 2003, 89.

---

6

This marked variation in supranational governance—variation that could be
demonstrated in regulatory regimes as well--is explained in part by the existence of
alternative modes of global governance. In the international economic governance,
including regulatory regimes, national governance within the boundaries of the territorial
state remains dominant. Under anarchy, states act unilaterally in pursuit of their interests
and rely upon ad hoc forms of cooperation buttressed by voluntary but self-enforcing
international regimes when interacting with others. This is the classic model of
international relations portrayed in neorealism or neoliberal institutionalism.8 When
national governance is insufficient, however, many have expected supranationalism to
replace it. Two alternative modes of governance at the international level—hierarchies
and networks--can substitute for both national governance and supranationalism,
however.

Supranationalism shifts political authority from individual states to states acting
as a collective body along two independent dimensions. States pool or share
sovereignty at the regional or global level by creating a collective unit that can make
binding or authoritative decisions for its members. The authority assigned to the
collective can vary in extent, covering more or fewer policy domains, and in
centralization, especially whether the collective power is subject to checks and balances
by other institutions and, importantly, by its member states (the equivalent of different
federal arrangements). States can also delegate authority to a supranational agent or
IGO. Delegation entails a conditional grant of authority from a state or collective of states
to an actor for some specified purpose. Most commonly, states delegate to
supranational agents in the forms of investigatory agents (e.g., IAEA) or dispute
resolution bodies (e.g., in the WTO), although other forms are possible.9 These two

8 See Waltz 1979 and Keohane 1984, respectively.
9 On delegation to IOs, see Hawkins, et al. 2006.

---

7
dimensions should be kept analytically separate. A decision to form a collective decision-
making unit is not a decision to delegate authority to an agent, or vice versa.10

Hierarchy shifts political authority from one or more subordinate states to a
dominant state that is empowered to issue binding decisions. The degree of hierarchy
varies by the number of policy issues over which a dominant state can legitimately
regulate the behavior of its subordinates. Also important are the residual rights of
control, or the ability to decide which party has authority over what when they disagree.
If the subordinate retains these residual rights of control, the relationship between
dominant and subordinate state is more anarchic, and if the dominant state acquires
these rights, the relationship is more hierarchic.11

In international relations, authority and, in turn, hierarchy rests not on a formal-
legal foundation but on an exchange relationship in which the dominant state provides a
social order or public goods of value to the subordinates, and subordinates in turn
recognize an obligation to comply with rules or resource claims that are necessary for
the production of that social order or public goods. In this way, hierarchy forms an
equilibrium that is regarded by both sides to the exchange as legitimate or appropriate.
For this equilibrium to endure, both the dominant and subordinate states must hold to
their sides of the implicit contract – the dominant state must provide the social order or
public goods, the subordinates must comply with its legitimate commands. The key
contracting difficulty under hierarchy is restraining opportunistic behavior by the
dominant state. Having yielded authority, subordinates correctly fear that the dominant
state will exploit them or seek to renegotiate the contract in its favor at some future
date.12

10 Lake and McCubbins 2006.
11 Lake 1999, forthcoming
12 Lake forthcoming

---

8

Perhaps the best illustration of hierarchy in the contemporary international
economy is dollarization, the hardest of exchange rate pegs, in which one country
adopts the currency of another as its own.13 In such cases, the dollarized economy
accepts the monetary authority of the country with the dominant currency. It does not
participate in decision-making regarding monetary policy in the dominant currency
country, and it exercises no influence on that policy. Authority runs in one direction, at
least in this domain. The authority relationship, even in this extreme example of
monetary hierarchy, is not a permanent transfer. The dollarized country can unwind its
adoption of the dominant currency—at considerable cost—and such relationships have
ended with some frequency, particularly when political hierarchies have been
disrupted.14 But within that relationship, the subordinate country imports the dominant
country’s monetary policy as its own.

Networks are enduring or repeated interactions among multiple actors that are
typically characterized by reciprocity. Distinguishing networks from alternative forms of
governance requires a narrowing of recent descriptions, which define networks as “a
pattern of regular and purposive relations among like government units working across
the border that divide countries from one another and that demarcate the ‘domestic’ from
the ‘international’ sphere.”15 Podolny and Page distinguish network from either hierarchy
or market, defining it as “any collection of actors (N≥2) that pursue repeated, enduring
exchange relations with one another and, at the same time, lack a legitimate
organizational authority to arbitrate and resolve disputes that may arise during the
exchange.” In contrast to markets or, in our case, national level governance, relations
between states in a network are enduring, not ad hoc. Networks are sometimes defined

13 Despite the name, dol arization applies to any such adoption, such as the use of the Euro as a
national currency by Montenegro or Kosovo.
14 On the costs and benefits of dol arization and near-dol arization, Cohen 2004, 123-139.
15 Slaughter 2004, 14.

---

9
as interactions in the absence of any authority, although it is more accurate to think of
them as sharing authority between the nodes or constituting authority from the
interactions themselves. If supranationalism implies delegation of authority to an IGO,
networked governance entails migration of authority to the policy network, but no formal
delegation. In contrast to hierarchies, recognized dispute settlement authority and
residual rights of control do not reside with any member of the network.16 If hierarchy in
international regulatory governance implies one-way transmission of templates with
authority residing in one national government, a networked structure implies reciprocity
(even if one member of the network has a more central position or more influence over
outcomes than others).17 Rather, the nodes form a collective of relatively equal units that
work together and share information to produce some common end.18 Networked
international governance, then, is based on shared or pooled authority and on repeated,
enduring, and reciprocal relationships among actors in different national jurisdictions.

Public-private and private networks were a familiar part of international
governance in the decades before 1945. In certain respects, the IGO-dominated
institutions after 1945 were created as substitutes for a form of networked governance
that was viewed as a bastion of unaccountable power and uncertain membership.
Transgovernmental networks did not disappear, however, and increasing economic
interdependence produced both renewed reliance on networked governance and a
scholarly rediscovery of those networks. Among transgovernmental networks, regulatory
networks of executive branch officials have received particular attention.19 Networked
regulatory governance often includes public-private networks or purely private networks

16 Podolny and Page 1998, 59.
17Walter Powell and others also add a distinctive ethic or norm that governs network relations,
particularly the norm of reciprocity and the consequent accretion of trust among its members.
(Powell 1990, 304-305)
18 Keck and Sikkink 1998. See also Kahler 2006, Lake and Wong 2006, and Eilstrup-Sangiovanni
2006.
19 Slaughter 2004, 36-64; Raustiala 2002.

---