Why Do Countries Choose to Globalize? Democracy- Versus Elite-Based Explanations Malcolm Fairbrother

Why Do Countries Choose to Globalize?
Democracy- Versus Elite-Based Explanations
Malcolm Fairbrother
School of Geographical Sciences
University of Bristol
University Road
Bristol BS8 1SS
United Kingdom
[email protected]
Tel: +44 117 9288303
October 2012
2 Why Do Countries Choose to Globalize?
Democracy- Versus Elite-Based Explanations
What drives the policy changes fostering economic globalization, in the sense of increasing
international trade and direct investment? A number of studies have found a statistical
association between economic openness and political democracy, and present globalization as a
consequence of states’ accountability to majority interests and preferences. Other research
challenges this view, however, emphasizing instead the preferences and power of elites. This
article evaluates these two perspectives by assessing evidence for the causal mechanisms
presumed by each of them. The weight of the evidence, including that derived from brief case
studies of Mexico and Canada, supports the elite-centered perspective more than the alternative,
calling into question the theory that democracy fosters economic globalization.
Keywords: globalization; democracy; elites; trade; NAFTA
3 Introduction
Globalization is a policy choice. Statistical estimates find that policy changes have been
the source of three-quarters of post-World War II growth in world trade as a share of GDP—
much more than technological innovation and diffusion (Baier and Bergstrand 2001). States have
often codified their decisions to pursue globalization in international agreements, which have had
large impacts on cross-border flows of trade and direct investment (Baier and Bergstrand 2007;
Büthe and Milner 2008; Goldstein, Rivers, and Tomz 2007; Neumayer and Spess 2005).
What, then, drives states’ decisions to open and/or to integrate their economies via trade
and/or direct investment?1 The literature on this question is sharply divided between arguments
emphasizing political democracy and others stressing the influence of economic, political, and/or
technical elites.
Evidence for a democracy-centered explanation of globalization comes from statistical
analyses finding cross-sectional and longitudinal associations between democracy and economic
openness. This perspective’s key theoretical premises are that, under democracy, the interests
and preferences of the majority rule, and globalization is in the interest of the majority. A recent
review of this literature, however, notes that the mechanisms linking presumed cause and effect
remain uncertain, and that there is a need for more work on precisely how democracy, arguably,
fosters globalization (Milner and Mukherjee 2009: 178; also Guisinger 2009: 554). In a literature
reliant on observational rather than experimental data, solid evidence for causal mechanisms
would seem important if not essential for demonstrating a causal relationship. This article
therefore assesses the plausibility of the mechanisms implied by a causal interpretation of the
statistical association between democracy and globalization.
4 To do so, the article makes use of Lieberman’s (2005) mixed-method strategy for
comparative research. For theory testing, he recommends “nested analyses” combining crossnational statistical modeling with studies of specific country cases conforming to fitted statistical
models. Such case studies test whether the experiences of specific countries are consistent with
the statistical associations uncovered by a large-N analysis. A valid theory should be capable of
predictions consistent not only with regressions across countries, but also causal processes within
them. This paper treats the cross-national regressions presented in other published studies as the
first stage of a nested analysis, and presents original case-based evidence as the second.
The alternative to the democracy-based approach is a set of arguments about the interests,
preferences, and power of elites, defined as any category of people in possession of relatively
large amounts of some valued resource, such as money, legal authority over public policy, or
recognized technical/intellectual expertise. Studies emphasizing the roles of economic, political,
and/or technical elites have suggested a more antithetical relationship between globalization and
democracy, wherein majority interests and/or preferences are not overpowering, but irrelevant or
even overridden. Studies comprising this second literature share a common emphasis on the
preferences and behaviors of elites, but are nonetheless heterogeneous, in that they concentrate
on different kinds of elites, with diverse relationships to the policymaking process, in countries
following a variety of distinct political economic pathways to globalization.2
To assess the democracy- and elite-centered perspectives, I first present generic evidence
about the causal mechanisms implied by each of them, and then evidence drawn from two
specific country cases: those of Mexico and Canada. The generic evidence consists of public
opinion data, insofar as democracy-based arguments rely heavily on models of individuals’
interest-based policy preferences. Mexico and Canada then serve as cases for brief qualitative
5 comparative-historical studies of why countries choose to globalize. These two countries
followed independent but parallel tracks to economic opening from the early 1980s to the mid1990s, culminating in their common enactment of the North American Free Trade Agreement
(NAFTA) in 1994. They both previously restrained their integration with the neighboring, much
larger United States economy; they both passed nationalist and restrictive foreign investment
laws in 1973; and in 1980 they jointly rejected U.S.-based proposals for continental economic
integration (Senate Standing Committee on Foreign Affairs 1980). In 1988, both countries held
elections that narrowly returned governments who subsequently dismantled their countries’ long
traditions of restraining rather than embracing international economic integration—particularly
on a regional basis. Considering their trajectories from the early 1980s to the mid-1990s thus
makes each country both a negative and a positive case (Haydu 1998); that fact, and their
spanning the developing and developed worlds, helps mitigate problems of selection bias.
To preview the paper’s main conclusions, both the generic and case-specific causal
process observations belie key implications of the theory that democracy fosters economic
globalization. In light of the lack of support for the theory’s posited mechanisms, I conclude that
the case for a causal relationship between democracy and globalization is not yet convincing,
despite the statistical association between them; the relationship may be spurious. The alternative
view that globalization is primarily a consequence of elite initiative is more consistent with the
observable causal mechanisms.
6 Globalization and its Drivers
A convincing explanation of globalization should be able to account for differences both
among countries with respect to their levels of integration, and among their trajectories over
time. According to one influential dataset, which codes the trade regimes of independent nations
as either “open” or “closed” for each year since 1950, only 37 out of 140 countries were open in
1980, and of these only 15 were developing countries.3 Ten years later, 61 countries were open,
and then 103 by 2000.4 With respect to direct investment and the multinationalization of
production, the stock of worldwide FDI as a share of GDP increased from 6.51 per cent in 1980
to 8.77 per cent in 1990, and then to 24.38 per cent by 2008.5
Most of the world was therefore globalizing only very slowly before the 1990s. Until
then, the United States and the Western European nations were generally consistent globalizers,
and small European nations were especially open (Katzenstein 1985). The smaller “western
offshoot” nations of Australia, Canada, and New Zealand were less enthusiastic, but still
generally open to the world economy. For the rest of the world, however, bracketing a few
outliers like the East Asian Tigers, only starting in the 1990s did trade and direct investment
begin to grow steadily, and rise above historical levels. While globalization’s prevalence in
recent times is widely known, the prevalence of its earlier absence, and the suddenness of the
“rush to free trade” in the late twentieth century (Rodrik 1994), are underappreciated. Many
developing countries, including large ones like Argentina or Brazil, were actually de-globalizing
for substantial periods of time. To illustrate how much globalization reversed previous trends, as
late as 1985 one influential scholar, in a book subtitled The Third World Against Global
Liberalism, named India and China as “developing countries that are substantially disengaged
7 from the world economy, and are likely to remain so” (Krasner 1985: 296). Until the 1980s,
there were international conflicts about international economic integration, and debates about
whether poorer countries stood to gain much from it (Frieden 2006). From World War II to the
1990s, globalization was, in short, limited to only a minority of (mostly wealthy) countries. Since
the 1990s, however, many have changed course dramatically.
Explanation I: Globalization and Democracy6
Coincident with the rise of economic liberalism in the late twentieth century, many
countries also transitioned to more politically liberal regimes in the 1980s and 1990s (see e.g.,
Schwartzman 1998; Wejnert 2005). And statistical analyses have found strong evidence that
democracy is positively associated with international economic liberalism.7 Democratization
tends to be followed by increases in trade flows, tariff reductions, and the liberalization of nontariff barriers to trade (Dutt and Mitra 2002; Eichengreen and Leblang 2008; Mansfield, Milner,
and Rosendorff 2002; see Milner and Mukherjee 2009 for a review). Pairs of democracies tend to
trade more than do other dyads (Mansfield, Milner and Rosendorff 2000). If democracy is taken
somewhat more abstractly to be relative equality of political power in society, there is other
evidence for the proposition that democracy and economic openness are associated. For example,
post-Communist countries with more fragmented state power have tended to liberalize trade
more than similar states where power has been more centralized—and this holds even for nondemocratic countries (Frye and Mansfield 2003). In short, the statistical evidence for an
association appears strong, though questions do remain about its robustness to alternative
8 measures of openness (Kono 2006) and the possibility that democracy is partly a consequence,
not just a cause, of globalization (Rudra 2005b).
Why should economic globalization be not just associated with, but caused by, political
democracy? Advocates of interpreting the statistical association as causal argue both that
democratization grants political power to a wider share of the population—making their policy
preferences more influential—and that most people will have favorable views of freer trade and
other kinds of economic liberalism because liberalism produces economic benefits for the
majority (Milner and Kubota 2005). This benign view of free trade rests on neoclassical
economic theory, which holds that trade allows countries to reap the gains of specialization and
comparative advantage, irrespective of the types of goods and services they produce (e.g.,
Krugman 1993). The end result is greater efficiency and the ability to consume more and/or
higher quality goods and services.
The theoretical rationale for linking globalization to democracy derives largely from the
expected distribution of globalization’s benefits among different constituencies in a country. The
neoclassical Heckscher-Ohlin-Samuelson model of trade (referred to hereafter as “HOS”)
predicts that a country’s relatively abundant factors of production stand to gain from
international economic integration, and scarce factors do not. A scarce factor, by virtue of its
being scarce, commands a high price in the status quo situation of economic closure, whereas
international integration makes it less scarce and thus undermines its power in the marketplace.
In developing countries contemplating integration with wealthier ones, low-skilled labor is the
relatively abundant factor, whereas in wealthier nations capital and highly skilled labor are the
abundant factors. Given this view of different people’s interests, and the rational choice
assumption that interests drive preferences, HOS predicts that capital owners and high-skill
9 workers in poorer countries will prefer economic closure, while lower-skilled workers will
support it. There is therefore theoretical reason to expect more liberalism in developing countries
where public policies are more accountable to majority opinion; autarky is likelier under
conditions of non-democracy, where political rulers cater to the domestic elite of capital owners
and highly skilled professionals (Milner and Mukherjee 2009). Governments’ responsiveness to
the will of the median voter, and the threat of electoral punishment, therefore drive decisions to
globalize; democracy fosters globalization by allowing the preferences of the majority to rule
(Busch and Mansfield 2010; Milner and Kubota 2005).
The democracy-centered literature tends to focus on developing countries, because of the
opportunities to exploit change over time in the level of a given country’s democracy; few
wealthy countries have recently transitioned to democracy. But the argument does have
implications for wealthier (and thus more capital-abundant and labor-scarce) nations. In such
contexts, HOS predicts the benefits of opening to flow more to capital owners than to workers
(Watson 1993). Yet because the median voter remains a worker, public opinion on globalization
should be mixed, at best (a prediction validated by the available survey data, as discussed
below). If democracy makes policymakers accountable to the preferences of the majority, and
the majority of voters in richer countries have ambiguous interests with respect to globalization,
then there is little reason to expect pro-globalization policies in such nations to derive from the
democratic accountability of politicians to the electorate. Consistent with this implication, one
recent study finds that while democracies and wealthier countries tend to be more open than nondemocracies and poorer countries, the interaction between these conditions is associated with
greater trade barriers (Tavares 2008).
10 In principle, democracy could lead to globalization not (or not just) because the public
supports it, but also for another reason. Democracy could foster globalization if it were to
increase the extent to which politicians are rewarded politically for the median voter’s prosperity,
even if public opinion or preferences were hostile or unresponsive. Voters might reward
decisions to increase international integration, even without being aware of it. This is not a
mechanism the democracy-based literature has addressed, however; instead, the emphasis has
been on the incentives confronting politicians in democratic versus authoritarian contexts,
assuming that there is majority support for globalization in each. This second, alternative
formulation of the democracy-centered theory makes strong assumptions about politicians’
foresight, understandings of economic policy, time horizons, and about the objective benefits of
globalization to the median voter on timescales relevant to the electoral cycle. And the available
evidence is not supportive of such a view, insofar as Kono (2008) finds that democracy is
associated with freer trade—but only in interaction with public support; where public support is
absent, democracy by itself is not associated with systematically more liberal trade policies.
Finally, the relationship between democracy and globalization could be spurious, rather
than causal, such as if both democratization and globalization were more likely to occur in
countries with greater exposure to a world culture sympathetic to political and economic
liberalism (Finnemore 1996). One key means by which world culture is transmitted is epistemic
communities, or groups of recognized experts (Simmons, Dobbin, and Garrett 2006). The
relative population of such experts varies from country to country, as does experts’ freedom to
travel or study abroad, and thereby to absorb norms and ideas which they may import back
home. It would not be surprising if the same non-democracies that came to be governed by
technocrats—highly educated individuals exposed to foreign expertise and ideas—were also
11 more likely to transition to democracy. The highly educated tend to be strong supporters of both
state accountability and free trade (Hainmueller and Hiscox 2006), such that while technocracy
only emerges in non-democratic contexts, many technocrats may nonetheless believe in the value
of democracy, and be more likely to introduce it.8 Once they gain power, and perhaps implement
some of their highest-priority economic policy changes, they may take steps to transition to
democracy. Or they may at least be less likely to use violence in repressing popular demands for
democracy, setting technocratic non-democracies apart from other, more intellectually and
economically autarkic nations. It is because of this possibility, at least, that the statistical
association between democracy and globalization is by itself not conclusive evidence of a causal
Explanation II: Globalization and the Power of Elites
In contrast to studies focused on democracy and the preferences of the majority, other
research emphasizes the agendas and actions of elites. Few studies argue explicitly that free
market policies have been implemented through undemocratic means, or that political democracy
is antithetical to free trade—though some do so (Harvey 2005). But the emphasis on the
behaviors and/or interests of the few, rather than the preferences and interests of the many, sets
this second literature at odds with the first. Elite-centered arguments are heterogeneous, and
emphasize at least four distinct categories of actors who they claim perform key roles in
fostering globalization. Each set of elites possesses a distinct relationship to the policymaking
process, given its unique source of power. This section discusses each category in turn.
12 First, one major strand of the political economy literature stresses the power of wellresourced international financial institutions and private creditors, and emphasizes their
preference for market-conforming policies (McMichael 2005; Park, Jang, and Lee 2007; Woods
2006). When a state becomes dependent on the IFIs (the World Bank and International Monetary
Fund), this dependence allows the latter to dictate policies to the former—such policies typically
being liberal and conducive to economic opening and integration. Dependence arises from debt
or balance of payments crises, where a developing state cannot repay its foreign creditors, and so
desperately requires short-term financial assistance. In another, somewhat softer variant of this
argument, dependence on the IFIs shifts the balance of power in domestic politics, and gives an
advantage to actors sympathetic to the same policies as the staff of the IFIs (e.g., Schneider
1998; Teichman 2004). Political actors with the training and connections that allow them to
negotiate with a state’s foreign creditors gain an advantage in competing for policymaking
control, and are thus eventually empowered to implement market-oriented policy revolutions as
an “inside job” (Babb 2001). A final variant suggests that the influence of the IFIs can be
intellectual: financial dependence makes it harder for states to ignore the abundance of IFIgenerated reports and studies advocating economic liberalism (Broad 2006; Edwards 1997).
A second category of explanations looks not to external constituencies, but to a key
domestic one: the private sector. Many studies present pro-globalization policy changes as the
product of proactive lobbying and campaigning by business, especially big business and/or
multinational firms specifically (Carroll 2004; Robinson 2004; Rupert 2000; Sklair 2002; Van
Apeldoorn 2002). This perspective holds that business wins from globalization, often at the
expense of other interest groups, and presumes that business unity in support of pro-globalization
policy changes is high (Harvey 2005).9 Work in this vein often adopts a very class-oriented view
13 of market-oriented policy changes generally, contrary to work rejecting that business is a unified
collective actor (e.g., Hart 2004). Pro-globalization agreements and policies benefit business by
helping to suppress wage and benefit demands by labor, and/or by entrenching pro-business
governance measures in areas like investor and/or intellectual property rights (Drahos and
Braithwaite 2002). Some statistical research has also emphasized business collective action,
showing for instance that in the U.S. congressional votes on NAFTA (in 1993) and the GATT
(1994) capital and labor groups were bitterly opposed, with business in favor (Beaulieu and
Magee 2004).
The third approach shifts to another domestic but extra-state constituency: economists.
Some literature suggests that recognized technical authorities in globalization-relevant policy
areas can use the force of their expertise to convince political elites and/or the general public to
become favorable to pro-globalization policies (Rupert 2000; Sheppard 2005). Economists in
universities and think tanks may have no formal policymaking authority, nor exceptional
financial resources at their disposal, but they have intellectual authority with respect to economic
policymaking, and thus arguably substantial influence over it (Bockman and Eyal 2002;
Montecinos and Markoff 2001). This authority increases the impact of their written reports, oral
testimony, and statements in the media, and gives them some influence over the views of noneconomists in policy-relevant roles—business, law, public administration, and the like. Surveybased research into economists’ policy preferences confirms that they are strongly unified in
support of liberal trade policies (Blendon et al. 1997; Frey et al. 1984). The neoclassical
framework has long subscribed to the idea of comparative advantage, and rejected mercantilism
as internally contradictory (Krugman 1993). Among industrial democracies since the Great
Depression, the strength of economists’ support for free trade has been consistently strong,
14 though support for globalization among developing country economists has been much more
variable, as the discipline of economics has changed substantially over the course of several
decades in many developing countries (Coats 1997; Babb 2001; Montecinos and Markoff 2001).
As developing country economists have become more similar to their developed country
counterparts (i.e., more neoclassical), their policy preferences have converged. One recent study,
albeit of another policy domain, therefore finds that the higher the number of American-trained
economists in a country, the higher the likelihood of privatization, consistent with the thesis that
the preferences of economists carry substantial weight (Kogut and Macpherson 2008). In short,
the number and policy preferences of the economists in a country can be important.
The fourth and final body of literature looks again at economists, but this time within
rather than outside the state. “Technocrats” are the final group of actors sometimes said to be key
agents of globalization (for discussions see Centeno 1993; Domínguez 1997; Markoff and
Montecinos 1993). Technocrats are state actors with formal authority over policymaking, which
allows their preferences to exercise significant direct influence over key outcomes. What
distinguishes technocrats from other politicians is their prior academic training and possession of
recognized credentials in economics. Such training makes them intellectually committed to proglobalization policies, to a much greater degree than other politicians would be, given specified
levels of support versus opposition by the general public and relevant interest groups. Nontechnocratic politicians may therefore bend with the political winds on trade—their policy
priorities are exogenously determined by interest group pressures. In contrast, technocrats, like
economists outside the state, have endogenous policy agendas conditioned by their own
biographies and training. Unlike economists in universities and think tanks, however, technocrats
have direct, formal authority over policy. As evidence for the influence of technocrats,
15 Chwieroth (2007) finds that having economists in top government posts increases the probability
of a country liberalizing its capital account.
In sum, these arguments all suggest that globalization follows from some category of
elites favoring international economic opening, acting in pursuit of that policy goal, and
possessing sufficient power such that they can indeed prompt a significant shift in public policy,
perhaps even in the face of significant opposition. Despite this commonality, the elite-centered
literature is heterogeneous; the arguments do vary somewhat. This is partly a consequence of the
fact that they address the circumstances of a diversity of countries. Arguments about the power
of the international financial institutions and about technocracy are more relevant for developing
countries, for example, whereas arguments about the influence of the domestic private sector are
more common among studies of wealthy industrialized nations.10
On the other hand, the various studies making up this literature are sometimes
contradictory. Some suggest for example that domestic technocrats are effectively the causal
mechanism by which the IFIs have their effects (e.g., Babb 2001), while others argue that IFI
engagement does not increase the probability of having technocratic leaders (Chwieroth 2007).
And few of these studies have seriously confronted the thesis that globalization has arisen largely
with democratization; nor have they confronted the international trade literature from which
arguments about democracy have been derived theoretically. This paper is the first to use case
studies of specific countries to assess both sets of arguments simultaneously.
16 Causal Processes Evidence
This section tests the democracy- and elite-centered theories against each other, by
examining the evidence for the observable implications of the causal mechanisms presumed by
each of them. Process-tracing can contribute to theory testing by checking for a previously
unspecified chain of events and/or conditions plausibly linking an independent variable or
antecedent condition to a dependent variable or outcome of interest (George and Bennett 2005;
Bennett and Elman 2007). The absence of such a chain suggests that a statistical association
between two variables may well be spurious.
The section begins by presenting broad-based evidence about causal processes in many
countries, and then proceeds to discuss the details of causal processes of specific cases.
First, the democracy-based explanation of globalization posits that public opinion
supports free trade in most countries. Is this view justified? The third wave of the World Values
Survey (conducted in the mid-1990s) asked nationally representative samples of respondents in a
diverse cross-section of 52 countries: “Do you think it is better if: (A) goods made in other
countries can be imported and sold here if people want to buy them, or (B) there should be
stricter limits on selling foreign goods here, to protect the jobs of people in this country?” In only
eight countries did more respondents answer A than B. Economists might point out that this is a
misleading question to ask people, insofar as trade actually has little impact on overall
employment in most countries (Krugman 1993). That, however, is beside the point under
consideration here: if people believe that trade depresses employment, and they support trade
restrictions more than trade opening as a consequence, the fact remains that economic
globalization is not a policy rooted in the preferences of the majority.
17 More recently, a Eurobarometer survey asked nationally representative samples of
Europeans whether they see themselves as “currently benefiting from international trade or not,”
with trade framed as imports/exports of goods and services into or out of the EU as a whole
(European Commission 2010). Including those who responded “Don’t Know,” 44 per cent of all
respondents said that they are benefiting from international trade, whereas 39 per cent said they
are not; only in six countries out of 28 did a majority or plurality say they were not benefiting.
This result could be taken as evidence of majority support for freer trade. However, the question
is ambiguous: is it about any benefits at all, or benefits net of any costs? A respondent could take
it as the former, and respond affirmatively to the question asked, even while viewing trade
negatively overall. Viewed in this way, the result is not evidence of great public enthusiasm.
Second, arguments about the consequences of political democracy for policy outcomes
rest heavily on the HOS trade model, described above. The theory predicts that elites in poorer
countries will favor autarky, while the majority will be more pro-trade, while this cleavage
should be reversed in richer countries. But are these predicted preferences actually supported by
the available survey data? In the case of the capital-abundant United States, there is indeed a
strong positive correlation between various measures of skill (or capital) and support for free
trade, as HOS would predict (Hainmueller and Hiscox 2006; Scheve and Slaughter 2001). And
cross-nationally, the more capital-abundant the country, the greater the impact of a worker’s
level of (human) capital on his/her support for free trade, another finding consistent with HOS
(Mayda and Rodrik 2005). There is not, however, the expected negative correlation between skill
and pro-trade sentiment in poorer countries: skilled workers favor freer trade in many countries
where HOS predicts they should not (Baker 2003). And analyses of individual-level survey data
belie a Heckscher-Ohlin-Samuelson-based model of people’s policy preferences fails in other
18 ways too. Some studies find that, contrary to HOS, policy preferences and political cleavages
follow the logic of a sectoral model known as Ricardo-Viner, wherein capital and labor are allied
within industries and divided across them (Hiscox 2001). Cross-sectionally, individuals’ trade
policy preferences appear related to their consumption, not just their positions in the labor
market (Baker 2005). And recent work based on U.S. data shows that trade as an issue has very
low salience for the mass public, such that voters do not use the ballot box to hold legislators
accountable to what trade policy preferences they do possess (Guisinger 2009). There is
therefore only partial support at most for predictions of people’s trade policy preferences based
on the trade model underlying the arguments about the influence on democracy.
Case-Specific Evidence about Causal Processes
I now turn to causal process observations specific to particular cases of economic closure
and opening. In order to select appropriate cases, I follow the recommendations of Lieberman’s
(2005) nested analysis approach. According to that approach, in instances where the results of an
initial large-N analysis appear broadly satisfactory, cases selected for subsequent qualitative
investigation should fall “on the [regression] line.” The rationale is that cases on the line should
possess causal processes that are more representative than exceptional. The case studies that
follow have thus been chosen because they fit the pattern identified in the large-N democracycentered literature: nations pursuing globalization in a context of political democracy.
I discuss two country cases—Mexico and Canada, a developing country that transitioned
to democracy at the end of the twentieth century, and a stably wealthy democracy. Each country
can be treated as at least two cases—one before and one after a significant episode of economic
19 opening—allowing for contrasts between positive and negative cases of globalization, and
avoiding problems of selection bias.11 Developing country cases that fall “on the line” for the
democracy-based literature are those where democratization preceded the enactment of proglobalization policies. The experiences of wealthier, more developed countries such as Canada
are not typically the focus of the democracy-based literature, much of which considers only
developing countries. But the theory that democracy fosters economic openness has implications
for wealthy democracies: their openness should be due, in substantial part, to their politicians’
accountability to publics that support free trade. In contrast, the elite-centered view charges that
the durability, if not expansion, of economic openness in any country is due to advocacy or
pressure from the powerful few, not the majority.
The accounts that follow are based on a broader comparative project about the political
economy of North American free trade generally, based on 114 interviews conducted by the
author with informants in Mexico, Canada, and the U.S. (mostly negotiators, other public
officials, politicians, political staff, and private sector representatives); analyses of a wide range
of archival evidence (governmental and private sector publications, transcripts of legislative
debates, news coverage, and public opinion polls); and previous academic studies and
journalistic accounts of North American free trade. Because interviewees agreed to speak on
condition that they not be quoted by name, and in the interest of saving space, the following
accounts do not include direct quotes; but claims based specifically on original interview data are
identified where appropriate.
20 Mexico
Prior to the 1980s, Mexico employed a strongly nationalist-statist, though still broadly
market-based, development model. From the mid-twentieth century through the early 1980s,
Mexico’s economic openness did not increase. Mexico declined to join the multilateral trading
system (the GATT), and relied on a complex system of tariffs, quotas, regulations, and import
licenses, to restrain its international economic integration via trade and investment (Lustig 1998).
In contrast, starting in the later 1980s and through the 1990s and 2000s, Mexico firmly embraced
international trade and investment, and saw dramatic increases in its international economic
integration. In 1986 it joined the GATT and in 1987 the state unilaterally accelerated its
international economic policy liberalization program, in an effort to deal with a surge of
inflation. According to the Wacziarg and Welch dataset, Mexico was open starting in 1987. But
Mexico’s most important international economic policy shift of the late twentieth century was
yet to come: the initiation of talks with the U.S. in 1990 about a free trade agreement (FTA). On
January 1st, 1994, as the culmination of trilateral negotiations among Mexico, the U.S., and
Canada, the North American Free Trade Agreement (NAFTA) entered into force. Over the
course of the 1990s, Mexico built on NAFTA in negotiating a series of bilateral and regional
trade agreements with other countries in both the developing and developed worlds.
Mexico’s economic opening followed its partial political liberalization in the 1980s and
early 1990s. In the minds of most casual observers, and according to the binary definition of
Przeworski et al. (2000), Mexico became a democracy in 2000. In that year, the ruling Partido
Revolucionario Institucional (Institutional Revolutionary Party, or PRI) lost the presidency for
21 the first time in decades. But, according to Marshall, Jaggers, and Gurr’s (2010) Polity IV
dataset, which treats democracy as gradational, there was a gradual increase in the country’s
level of democracy starting in the 1970s (see Figure 1). Mexican authoritarianism took the form
of a dominant party system, where elections were held and opposition parties were allowed to
campaign, but the PRI used state funds to finance patronage and disproportionate campaign
spending that consistently ensured its electoral success (Dresser 1991). The system gradually
broke down, as opposition parties became increasingly popular and influential, evidenced by a
steady decline in the PRI’s seat share in the lower house of the congress from 1961 to 2000
(Greene 2007), and the growing share of sub-national elections won by opposition parties. After
1982, conditions of economic crisis and scarcity further undermined the PRI’s financial
privilege. At the same time, secular socio-demographic and economic changes in Mexico
fostered the emergence of an increasingly autonomous and assertive mass media, important
elements of which were strong critics of the ruling regime (Lawson 2002). In 1988, the threat of
losing the presidential election for the first time in decades forced the PRI to resort to widely
decried fraud; whether the PRI candidate, Carlos Salinas de Gortari, really won a majority of the
votes cast in that election will never be known.
[Fig. 1 about here]
In short, the Mexican state’s democratic accountability was growing just at the time that
it deepened Mexico’s economic opening and integration into international markets, meaning
Mexico fits the pattern found in the democracy-centered literature. Were these two changes over
time causally related, however? Was economic opening due to the state’s growing accountability
22 to a mass public actively desirous of economic opening? Did democratization erode the policy
influence of privileged minorities who benefitted from closure?
A number of studies have shown that while large nationalized industries and many small
manufacturers were hostile to economic opening in the 1980s and 1990s, large private firms—
whose weight was growing in Mexico’s economy as a whole—were enthusiastic (Flores Quiroga
1998; Shadlen 2000; Thacker 2000). It is not the case, then, that democratization undermined the
power of a domestic economic elite that supported autarky: Mexico’s top businesspeople were
advocates of opening. Elements of the old (and import-competing) manufacturing sector fought
the opening, but they were not able to prevent all of the major national business associations,
such as the national manufacturers’ association Concamin, from eventually endorsing NAFTA
Conversely, there is little evidence of active pressure from the Mexican public for
economic opening in the 1980s or 1990s.12 That is not to say that most Mexicans were strongly
opposed. Some public opinion data—from 1990 onwards, regarding NAFTA specifically—
suggest that the majority of Mexicans generally accepted the idea of free trade with the U.S.13 On
the other hand, far more Mexican respondents to the 1996 World Values Survey favored “stricter
limits on selling foreign goods here, to protect the jobs of people in this country” than allowing
“goods made in other countries [to] be imported and sold here if people want to buy them” (76 to
16 per cent). Either way, the general public was little involved, and not much exercised about the
issue of freer trade. Democratic debate was constrained, and most Mexicans remained only
aware of but not actively engaged in that debate.
Public opinion therefore was not a major consideration for the officials deciding on
Mexico’s international trade and investment policies. With respect to the decisions to accede to
23 GATT and to pursue NAFTA, American officials had long made clear that the U.S. would
welcome a closer economic relationship with Mexico, including in the form of a free trade zone.
When President Salinas instructed his staff to broach the subject of an FTA with their American
counterparts in early 1990, they feared a public backlash far more than they expected public
enthusiasm, and they made no announcement about the talks until a journalist uncovered the
story. The decision to seek an agreement was aimed at attracting more foreign investment to
Mexico, and facilitating market-oriented policy changes that many Mexican officials would have
been willing to make even unilaterally. The decisions to propose a free trade agreement with the
U.S., to negotiate it, and to ratify it were all largely orthogonal to public opinion.14
Instead of being the result of the state’s increasing democratic accountability to a clearly
trade-supportive public, case-based accounts of Mexico’s opening are more consistent with some
of the elite-based arguments outlined earlier. Starting in the early 1980s, economic hard times
made Mexico dependent on the World Bank, IMF, and private external creditors. Over the
course of the decade, the upper echelons of the Mexican state became populated by U.S.-trained
economists with distinctly market-oriented economic ideas. This new generation of policymakers
gradually displaced their older, more statist-nationalist competitors, to the point where much of
the federal cabinet consistent of economists with PhDs from Chicago, Yale, MIT, and the like
(see Babb 2001; Centeno 1997; Thacker 2000). This combination of external financial
constraints and internal activism by true-believing technocrats led to a series of pro-globalization
policy changes that undid Mexico’s inward-oriented industrialization model (interviews). The
policy preferences of the new national political elite reflected the biographies of key office
24 Canada
For Canada, even more than for Mexico, the late-twentieth century embrace of
globalization was centered on the negotiation of a free trade agreement with the United States.
Like Mexico, Canada had historically sought to restrain, rather than deepen, its relationship with
its much larger neighbor. But once it stopped rejecting the idea of bilateral free trade, foreign
trade and investment expanded rapidly relative to GDP. The Canada-U.S. bilateral free trade
agreement (CUFTA) went into effect at the start of 1989, five years before the trilateral NAFTA.
From a Canadian perspective, NAFTA was largely an extension and expansion of the earlier
bilateral agreement, and it was CUFTA that embodied an enormously controversial policy
change in Canada. The public debate about it in the late 1980s was intense, and arguably more so
than for any other international economic policy decision in Canada’s history.
Was Canadian public opinion, and the state’s accountability to it, a key driver of the
Canadian government’s decision to seek free trade with the U.S. (and later with Mexico)?
In the short term, CUFTA’s passage came to depend on the re-election of the incumbent
Progressive Conservative (“Tory”) government that negotiated the agreement between 1985 and
1987. In the federal parliamentary election of 1988, only the Tories promised to enact the
agreement if elected; the other two major parties (the Liberals and New Democrats) made their
opposition central to their platforms. As the campaign unfolded, the free trade agreement became
the main debate and issue—a rare instance where trade became a salient public concern. In the
end, the Tories were successfully re-elected, with more than half the seats in the parliament—but
only a plurality (43 per cent) of the popular vote. The centrist Liberals and labor-allied New
Democrats received 32 and 20 per cent of the popular vote respectively: a majority, if combined.
25 There is good reason to think that these vote shares derived in substantial part from people’s
opinions of free trade: a poll taken in the same month as the election found 39 per cent support
and 51 per cent opposition to the FTA (LeDuc 1989). The evidence about public opinion does
not therefore suggest that the Canadian electorate supported the principle policy change by
which Canada’s economy was globalized in the late twentieth century. CUFTA would not have
passed had public opinion alone determined its fate.
During the negotiation and ratification of NAFTA in the early 1990s, public opinion in
Canada turned even more negative about free trade. The Canadian economy went into recession
just as it was beginning to adjust to new market conditions under CUFTA. In 1993, 33 per cent
of respondents to one survey reported supporting free trade, while 60 per cent were opposed;
regarding the then still-to-be-enacted NAFTA, 23 per cent were in favor and 69 per cent were
opposed.15 Despite the popular hostility to free trade, however, a new Liberal government elected
in the fall of 1993 enacted NAFTA on schedule. By that time, the Liberals had quietly dropped
their opposition to free trade, the new party leader having in 1991 called protectionism “passé”
and globalization “a fact of life” (Chrétien 1991, quoted in Windsor 1991). During the 1993
campaign, recognizing the public’s hostility to NAFTA, the Liberals promised to renegotiate the
agreement if they were elected, and add provisions regarding subsidies, dumping, dispute
settlement, and energy (Alexandroff 1993: 57). Once elected, however, the changes they made
were cosmetic, at most; the international negotiations lasted all of two days, and resulted only in
unilateral statements by the Canadian government and agreement in principle to continue
discussions on some topics by the U.S. Much more than in Mexico, then, there is reason to think
that even had the opposition won in Canada’s 1988 election, the federal government would soon
26 have ended up enacting international economic policies much like those of the incumbent
governing party.
If the Canadian state’s democratic accountability cannot explain its support for increasing
international economic openness in the late twentieth century, what can? Like for Mexico, causal
process observations from the Canadian case are consistent with one of the elite-centered
arguments discussed above—though in the Canadian case, the originating advocates of a major
pro-globalization policy shift were based in the private rather than public sector.
Canada’s post-World War II international economic policies were broadly liberal, but
less so than the median developed country.16 While many Canadian bureaucrats and politicians
were sympathetic to economists’ arguments for free trade—and leading academic economists in
Canada were strongly supportive of free trade both generally and with the U.S. specifically
(Economic Council of Canada 1975; Johnson 1968)—Canadian business was not. The result was
a sometimes hostile standoff between the private and public sectors over the issue of Canada’s
economic openness, with opposition to freer trade being particularly rooted in the manufacturing
sector. Only in the early 1980s did Canadian manufacturers became much more sanguine about
their ability to thrive in international markets, and with their reversal of views, the Canadian
private sector as a whole became very motivated to pursue a bilateral Canada-U.S. free trade
agreement (Langille 1987; interviews). Between the late 1970s and the mid-1980s, the
preferences of Canadian business changed substantially. In 1977, the Canadian Manufacturers’
Association (CMA) stated that “at this stage in Canada’s development as an industrial nation, our
industry is in general not as capable of benefiting from trade concessions gained as it is
vulnerable to trade concessions granted.” By 1984, the CMA had substantially changed its tune,
saying “Canadian industry needs further multilateral trade liberalization… Moreover, it needs to
27 [give] consideration to entering into a trade agreement with the U.S.” Although some previous
studies have found sectoral divides with respect to trade and globalization, starting in the mid1980s the business community in Canada became strongly united in favor of greater opening,
with the few dissenters being relatively small industries (e.g., furniture, wine).
The Tory government’s mid-1985 decision to pursue free trade with the U.S. followed
very soon after these key changes in the trade policy preferences of the Canadian business
community. In 1985, the government accepted a long-standing tacit invitation from the U.S. to
negotiate. The decision was taken after extensive consultations with horizontal and industryspecific business associations (Simeon 1987), and the government’s priorities in the subsequent
negotiations—above all secure export access to the U.S. market—were heavily driven by private
sector priorities. Business campaigning for free trade was extensive before, during, and after the
negotiations, and dovetailed with a long-standing enthusiasm for the idea on the part of liberalminded bureaucrats in Ottawa, as well as outside economic policy experts in Canadian
universities and think tanks (Golob 2003). The Tory government had numerous ties to the private
sector, shared many of its views, and generally sought to pursue business priorities as a matter of
course (Watson 1987). In sum, economists’ views about the desirability of free trade did not
change over time, bureaucrats’ views changed little, the views of business changed a great deal,
and the priorities of politicians followed those of Canadian business. The pattern of changes over
time differed substantially from the pattern in Mexico, where business views evolved little, and
the priorities of politicians and top bureaucrats evolved a lot, as previous generations of political
elites were replaced by technocrats.
The influence and policy preferences of the Canadian private sector are further
demonstrated by the intensity and sectoral breadth of the business campaigning for free trade
28 between 1986 and the 1988, and during the 1988 election campaign for the pro-free trade Tories.
The country’s three major business associations—the Canadian Manufacturers’ Association,
Canadian Chamber of Commerce, and the Business Council on National Issues—joined together
to campaign strongly in support of free trade, through an alliance they called CATJO (“Canadian
Alliance for Trade and Job Opportunities”—see Ayres 1998; Doern and Tomlin 1991). Without
CATJO’s support, CUFTA might well have failed, because the Tories would have received
fewer votes, making collective action by business a critical foundation for Canada’s
globalization. As the Prime Minister’s then-Chief of Staff writes in his memoirs, in 1988 “the
business community rallied in unprecedented fashion to demonstrate support for the agreement”
(Burney 2005).
Both the generic and the case-specific causal process observations presented in this paper
belie the proposition that mass public opinion and/or voters’ demands drive the policy changes
that open national economies to international markets. Public opinion surveys find no more than
mixed evidence about people’s support for free trade; the highly educated are consistently
supportive but other people are not, and people’s international economic policy preferences do
not appear to make much difference when they decide who to vote for. In the specific cases of
Mexico and Canada, political elites pursued pro-globalization policies for reasons that had little
to do with public opinion or their democratic accountability to it, and in some respects public
opinion weighed against North American free trade.
29 The evidence for a causal relationship between democracy and globalization is therefore
questionable, despite the statistical association between them. The association could be due to
the presence of some common driver of both democratization and globalization; as discussed
earlier, one possibility is that the rise of certain kinds of political elites in authoritarian states,
such as technocrats with substantial previous exposure to world culture, may create the
conditions for both pro-globalization policy changes and subsequent transitions to democracy.
Researchers who find statistical associations between democracy and globalization may
not intend to make completely causal arguments, but the ways in which this association is
presented tend to lead to a causal interpretation. This paper has shown that such an interpretation
is not warranted. Instead, arguments about the power of elites are more convincing—though the
relevant elites vary across national contexts, given the important political economic differences
between developing and developed countries in particular.
Alexandroff AS (1993) The Federal and Provincial Liberal Parties and NAFTA. In: Riggs AR
and Velk R (eds) Beyond NAFTA: An Economic, Political and Sociological Perspective.
Vancouver: Fraser Institute, 56-62.
Ayres JM (1998) Defying Conventional Wisdom: Political Movements and Popular Contention
against North American Free Trade. Toronto: University of Toronto Press.
Babb S (2001) Managing Mexico: Economists from Nationalism to Neoliberalism. Princeton,
NJ: Princeton University Press.
30 Baier SL and Bergstrand JH (2001) The Growth of World Trade: Tariffs, Transport Costs, and
Income Similarity. Journal of International Economics 53: 1-27.
Baier SL and Bergstrand JH (2007) Do Free Trade Agreements Actually Increase Members’
International Trade? Journal of International Economics 71: 72-95.
Baker A (2003) Why Is Trade Reform So Popular in Latin America? A Consumption-Based
Theory of Trade Policy Preferences. World Politics 55: 423-55.
Baker A (2005) Who Wants to Globalize? Consumer Tastes and Labor Markets in a Theory of
Trade Policy Beliefs. American Journal of Political Science 49: 924-38.
Beaulieu E and Magee C (2004) Four Simple Tests of Campaign Contributions and Trade Policy
Preferences. Economics & Politics 16: 163-87.
Bennett A and Elman C (2007) Case Study Methods in the International Relations Subfield.
Comparative Political Studies 40: 170-95.
Bernhard M and Karakoç E (2007) Civil Society and the Legacies of Dictatorship. World
Politics 59: 539–67
Blendon RJ, Benson JM, Brodie M, Morin R, Altman DE, Gitterman D, Brossard M and James
M (1997) Bridging the Gap Between the Public’s and Economists’ Views of the
Economy. Journal of Economic Perspectives 11: 105-18.
Bockman J and Eyal G (2002) Eastern Europe as a Laboratory for Economic Knowledge: The
Transnational Roots of Neoliberalism. American Journal of Sociology 108: 310-52.
Broad R (2006) Research, Knowledge, and the Art of ‘Paradigm Maintenance’: The World
Bank’s Development Economics Vice-Presidency (DEC). Review of International
Political Economy 13: 387-419.
31 Burney DH (2005) Getting It Done: A Memoir. Montreal and Kingston: McGill-Queen’s
University Press.
Busch ML and Mansfield ED (2010) Trade: Determinants of Policies. In: Denemark RA et al.,
editors. International Studies Compendium Project. Oxford: Wiley-Blackwell.
Busch ML and Reinhardt E (2005) Industrial Location and Voter Participation in Europe. British
Journal of Political Science 35: 713-30.
Büthe T and Milner HV (2008) The Politics of Foreign Direct Investment into Developing
Countries: Increasing FDI through International Trade Agreements? American Journal of
Political Science 52: 741-62.
Canadian Manufacturers’ Association (1977) Agenda for Action. Available from Library and
Archives Canada, Royal Commission on the Economic Union and Development
Prospects for Canada, Volume 29, File: #79, Brief Related Material, Brief No. 79.
Canadian Manufacturers’ Association (1984) A Future that Works: The Canadian
Manufacturers’ Association Outlines a Possible Future and How to Get There. Toronto:
Canadian Manufacturers’ Association. Available from Library and Archives Canada,
Amicus 4894340.
Carroll WK (2004) Corporate Power in a Globalizing World: A Study in Elite Social
Organization. Don Mills, ON: Oxford University Press.
Centeno MÁ (1993) The New Leviathan: The Dynamics and Limits of Technocracy. Theory and
Society 22: 307-35.
Centeno MÁ (1997) Democracy Within Reason: Technocratic Revolution in Mexico. 2nd ed.
University Park, PA: Pennsylvania State University Press.
32 Chrétien J (1991) Closing remarks delivered at the Liberal Party of Canada Conference, Aylmer,
Québec, November 24 1991. Quoted in Hugh Windsor. Liberals Conference Tests Icy
Economic Waters. Globe and Mail, 26 November; A1 and A8.
Chwieroth J (2007 Neoliberal Economists and Capital Account Liberalization in Emerging
Markets. International Organization 61: 443-63.
Coats AW (ed, 1997) The Post-1945 Internationalization of Economics. Durham: Duke
University Press.
Doern B and Tomlin BW (1991) Faith and Fear: The Free Trade Story. Toronto: Stoddart.
Domínguez JI (ed, 1997) Technopols: Freeing Politics and Markets in Latin America in the
1990s. University Park, PA: Pennsylvania State University Press.
Drahos P and Braithwaite J (2002) Information Feudalism: Who Owns the Knowledge Economy?
London: Earthscan.
Dreiling M (2001) Solidarity and Contention: The Politics of Security and Sustainability in the
NAFTA Conflict. New York: Garland.
Dresser D (1991) Neopopulist Solutions to Neoliberal Problems: Mexico’s National Solidarity
Program. San Diego: UCSD Center for U.S.-Mexican Studies.
Drope J (2006) Don’t Buy from Me Argentina: Politics, Economics, and Trade Liberalization in
Argentina, 1992-2001. Studies in Comparative International Development 41: 53-75.
Dutt P and Mitra D (2002) Endogenous Trade Policy Through Majority Voting: An Empirical
Investigation. Journal of International Economics 58: 107-33.
Economic Council of Canada (1975) Looking Outward: A New Trade Strategy for Canada.
Ottawa: Information Canada.
33 Edwards S (1997) Trade Liberalization Reforms and the World Bank. American Economic
Review 87: 43-8.
Ehrlich SD (2007) Access to Protection: Domestic Institutions and Trade Policy in Democracies.
International Organization 61: 571-605.
Eichengreen B and Leblang D (2008) Democracy and Globalization. Economics & Politics 20:
European Commission (2010) Eurobarometer 74.1: International Trade. Brussels: European
Commission Directorates General for Trade and Communication.
Finnemore M (1996) Norms, Culture, and World Politics: Insights from Sociology’s
Institutionalism. International Organization 50: 325-47.
Flores Quiroga AR (1998) Proteccionismo Versus Librecambio: La Economía Política de la
Protección Comercial en México, 1970-1994. México: Fondo de Cultura Económica.
Fourcade-Gourinchas M and Babb S (2002) The Rebirth of the Liberal Creed: Paths to
Neoliberalism in Four Countries. American Journal of Sociology 108: 533-79.
Frey BS, Pommerehne WW, Schneider F and Gilbert G (1984) Consensus and Dissension
Among Economists: An Empirical Inquiry. American Economic Review 74: 986-94.
Frieden JA (2006) Global Capitalism: Its Fall and Rise in the Twentieth Century. New York:
Frye T and Mansfield ED (2003) Concentration of Government Authority and Trade
Liberalization in Post-Communist Countries. British Journal of Political Science 33:
Garrett G (2000) The Causes of Globalization. Comparative Political Studies 33: 941-91.
34 George AL and Bennett A (2005) Case Studies and Theory Development in the Social Sciences.
Cambridge, MA: MIT Press.
Goldstein JL, Rivers D and Tomz M (2007) Institutions in International Relations:
Understanding the Effects of the GATT and the WTO on World Trade. International
Organization 61: 37-67.
Golob SR (2003) Beyond the Policy Frontier: Canada, Mexico, and the Ideological Origins of
NAFTA. World Politics 55: 361–98.
Greene KF (2007) Why Dominant Parties Lose: Mexico’s Democratization in Comparative
Perspective. New York: Cambridge University Press.
Hainmueller J and Hiscox MJ (2006) Learning to Love Globalization: Education and Individual
Attitudes Toward International Trade. International Organization 60: 469–98.
Hart DM (2004) ‘Business’ Is Not an Interest Group: On the Study of Companies in American
National Politics. Annual Review of Political Science 7: 47-69.
Hart M (2002) A Trading Nation: Canadian Trade Policy from Colonialism to Globalization.
Vancouver: University of British Columbia Press.
Harvey D (2005) A Brief History of Neoliberalism. Oxford: Oxford University Press.
Haydu J (1998) Making Use of the Past: Time Periods as Cases to Compare and as Sequences of
Problem Solving. American Journal of Sociology 104(2): 339-71.
Hiscox MJ (2001) Class Versus Industry Cleavages: Inter-Industry Factor Mobility and the
Politics of Trade. International Organization 55: 1-46.
Johnson HG (1968) Canadian Contributions to the Discipline of Economics Since 1945.
Canadian Journal of Economics 1: 129-46.
35 Katzenstein PJ (1985) Small States in World Markets: Industrial Policy in Europe. Ithaca, NY:
Cornell University Press.
Kogut B and Macpherson E (2008) The Decision to Privatize: Economists and the Construction
of Ideas and Policies. In: Simmons BA, Dobbin F and Garrett G (eds) The Global
Diffusion of Markets and Democracy. New York: Cambridge University Press, 104-40.
Kono DY (2006) Optimal Obfuscation: Democracy and Trade Policy Transparency. American
Political Science Review 100: 369-84.
Kono DY (2008) Does Public Opinion Affect Trade Policy? Business and Politics 10: Article 2.
Krasner SD (1985) Structural Conflict: The Third World Against Global Liberalism. Berkeley:
University of California Press.
Krugman PR (1993) What Do Undergrads Need to Know About Trade? American Economic
Review 83: 23-6.
Kus B (2006) Neoliberalism, Institutional Change and the Welfare State: The Case of Britain and
France. International Journal of Comparative Sociology 47(6): 488-525.
Langille D (1987) The Business Council on National Issues and the Canadian State. Studies in
Political Economy 24: 41-85.
Lawson CH (2002) Building the Fourth Estate: Democratization and the Rise of a Free Press in
Mexico. Berkeley: University of California Press.
LeDuc L (1989) The Canadian Federal Election of 1988. Electoral Studies 8: 163-67.
Lieberman ES (2005) Nested Analysis as a Mixed-Method Strategy for Comparative Research.
American Political Science Review 99: 435-52.
Lustig N (1998) Mexico: The Remaking of an Economy. 2nd ed. Washington DC: Brookings.
36 Mansfield ED, Milner HV and Rosendorff BP (2000) Free to Trade: Democracies, Autocracies,
and International Trade. American Political Science Review 94: 305-21.
Mansfield ED, Milner HV and Rosendorff BP (2002) Why Democracies Cooperate More:
Electoral Control and International Trade Agreements. International Organization 56:
Markoff J and Montecinos V (1993) The Ubiquitous Rise of Economists. Journal of Public
Policy 13: 37-68.
Marshall MG, Jaggers K, and Gurr TR (2010) Polity IV Project: Political Regime Characteristics
and Transitions, 1800-2009. Available: http://www.systemicpeace.org/polity/polity4.htm.
Mayda AM and Rodrik D (2005) Why Are Some People (and Countries) More Protectionist than
Others? European Economic Review 49: 1393-430.
McGillivray F (2004) Privileging Industry: The Comparative Politics of Trade and Industrial
Policy. Princeton: Princeton University Press.
McMichael J (2005) Globalization. In: Janoski T, Alford R, Hicks A and Schwartz MA (eds) The
Handbook of Political Sociology: States, Civil Societies, and Globalization. New York:
Cambridge University Press, 587-606.
Milner HV and Kubota K (2005) Why the Move to Free Trade? Democracy and Trade Policy in
the Developing Countries. International Organization 59: 107-43.
Milner HV and Mukherjee B (2009) Democratization and Economic Globalization. Annual
Review of Political Science 12: 163–81.
Montecinos V and Markoff J (2001) From the Power of Economic Ideas to the Power of
Economists. In: Centeno MA and López-Alves F (eds) The Other Mirror: Grand Theory
through the Lens of Latin America. Princeton NJ: Princeton University Press, 105-50.
37 Neumayer E and Spess L (2005) Do Bilateral Investment Treaties Increase Foreign Direct
Investment to Developing Countries? World Development 33: 1567-85.
Park GS, Jang YS and Lee HY (2007) The Interplay between Globalness and Localness: Korea’s
Globalization Revisited. International Journal of Comparative Sociology 48(4): 337-53.
Przeworski A, Alvarez ME, Cheibub JA and Limongi F (2000) Democracy and Development:
Political Institutions and Well-Being in the World, 1950-1990. New York: Cambridge
University Press.
Robinson WI (2004) A Theory of Global Capitalism: Production, Class, and State in a
Transnational World. Baltimore: Johns Hopkins University Press.
Rodrik D (1994) The Rush to Free Trade in the Developing World: Why So Late? Why Now?
Will It Last? In: Haggard S and Webb SB (eds) Voting for Reform: Democracy, Political
Liberalization, and Economic Adjustment. Oxford: Oxford University Press, 61–88.
Rudra N (2005a) Are Workers in the Developing World Winners or Losers in the Current Era of
Globalization? Studies in Comparative International Development 40: 29-64.
Rudra N (2005b) Globalization and the Strengthening of Democracy in the Developing World.
American Journal of Political Science 49: 704-30.
Rupert M (2000) Ideologies of Globalization: Contending Visions of a New World Order. New
York: Routledge.
Sachs JD and Warner A (1995) Economic Reform and the Process of Global Integration.
Brookings Papers on Economic Activity 1: 1-118.
Salinas de Gortari C (2000) México: Un Paso Difícil a la Modernidad. Barcelona: Plaza &
38 Scheve KF and Slaughter MJ (2001) Globalization and the Perceptions of American Workers.
Washington DC: Institute for International Economics.
Schneider BR (1998) The Material Bases of Technocracy: Investor Confidence and
Neoliberalism in Latin America. In: Centeno MA and Silva P (eds) The Politics of
Expertise in Latin America. New York: St. Martin’s, 77-95.
Schwartzman KC (1998) Globalization and Democracy. Annual Review of Sociology 24: 159-81.
Senate Standing Committee on Foreign Affairs (1980) Proceedings. Issue No. 3, June 3.
Shadlen KC (2000) Neoliberalism, Corporatism, and Small Business Political Activism in
Contemporary Mexico. Latin American Research Review 35: 73-106.
Sheppard E (2005) Constructing Free Trade: From Manchester Boosterism to Global
Management. Transactions of the Institute of British Geographers 30: 151-72.
Simeon R (1987) Inside the Macdonald Commission. Studies in Political Economy 22: 167-79.
Simmons BA, Dobbin F and Garrett G (2006) Introduction: The International Diffusion of
Liberalism. International Organization 60: 781-810.
Sklair L (2002) Globalization: Capitalism and its Alternatives. Oxford: Oxford University Press.
Tavares JA (2008) Trade, Factor Proportions and Politics. Review of Economics and Statistics
90: 163–68.
Teichman J (2004) The World Bank and Policy Reform in Mexico and Argentina. Latin
American Politics and Society 46: 39-74.
Thacker SC (2000) Big Business, the State, and Free Trade: Constructing Coalitions in Mexico.
New York: Cambridge University Press.
Van Apeldoorn B (2002) Transnational Capitalism and the Struggle over European Integration.
London: Routledge.
39 Wacziarg R and Welch KH (2008) Trade Liberalization and Growth: New Evidence. World
Bank Economic Review 22: 187-231.
Watson W (1993) The Economics of NAFTA: How Much Zero-Sum? In: Riggs AR and Velk T
(eds) Beyond NAFTA: An Economic, Political and Sociological Perspective. Vancouver:
Fraser Institute, 159-66.
Watson WG (1987) Canada-US Free Trade: Why Now? Canadian Public Policy 13: 337-49.
Wejnert B (2005) Diffusion, Development, and Democracy, 1800–1999. American Sociological
Review 70: 53-81.
Woods N (2006) The Globalizers: The IMF, the World Bank, and their Borrowers. Ithaca:
Cornell University Press.
40 10 8 6 4 2 0 -­2 -­4 -­6 -­8 Democracy 80 70 60 50 40 30 20 10 0 1950 1953 1956 1959 1962 1965 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 Trade/GDP (%) Figure 1: Mexican Democracy and Trade, 1950-­2007 (Sources: Penn World Table 6.3; Polity IV) Trade/GDP (%) Democracy 1
This article does not address the globalization of finance, which raises different issues and deserves
separate treatment; trade and direct investment are more closely related (see Garrett 2000).
Some arguments presented here will also be relevant for the rise of other kinds of market-oriented or
neoliberal economic policies in the late 20th century (see e.g., Fourcade-Gourinchas and Babb 2002; Kus
2006). But the political origins of domestic and international neoliberalism are to some extent distinct,
and not all explanations provided of neoliberal policies like tax cuts, welfare state retrenchment, and/or
privatization (for example) are covered.
The dataset, described and made available in Wacziarg and Welch (2008), is an updated and refined
version of one constructed by Sachs and Warner (1995).
The binary measurement of economic opening reflects that countries have tended to make transitions to
globalization at distinct and identifiable moments in their histories, such as through decisive actions like
acceding to GATT or negotiating and ratifying an agreement with one or more sizeable neighbors. Binary
classifications are not perfect, but neither are other possible indices (Milner and Kubota 2005: 122-3).
41 Tariff rates alone do not reflect important non-tariff barriers (Drope 2006; Kono 2006), and it is not clear
how one could combine tariffs and non-tariff barriers into a single summative index.
These figures come from UNCTAD’s Foreign Direct Investment on-line database.
Studies of trade and globalization have also pointed to a variety of influences not otherwise addressed
here, including domestic electoral institutions, party cleavages, and economic geography (see for example
Busch and Reinhardt 2005; Drope 2006; Ehrlich 2007; McGillivray 2004; for a review see Busch and
Mansfield 2010). These other factors have however been used much more to explain differences across
the treatment of industries within a given country, or differences across countries, at a given point in time.
As explanations of change over time within countries, existing literature has placed primary emphasis on
the influence of political democratization, or of elite action.
Many studies do not define provide formal definitions of democracy, but simply work with measures
from large datasets like Polity IV and Freedom House, which concentrate on procedural and institutional
criteria like the existence of elections and constraints on the executive, or on the protection in practice of
political rights. I adopt this approach here as well.
Bernhard and Karakoç (2007) find that education is positively associated with participation in both
organizational life and political protests.
Consistent with this view, for example, Rudra (2005a) finds that the bargaining power of labor has
declined with globalization across most of the developing world.
The latter difference is consistent with expectations derived from the Heckscher-Ohlin-Samuelson
model of trade, discussed earlier, which implies that developing country capitalists will be less
enthusiastic about globalization, because it will expose them to fierce international competition and drive
down their profits.
Canada was a broadly open economy even before the 1980s. But the stark differences between its
policies before and after the 1980s make for a useful comparison with cases of transition in developing
An important exception in this regard may be for labor mobility.
42 13
As a sign of the state’s growing sense of democratic accountability, the executive branch grew more
concerned about public opinion, and in the early 1990s the Salinas administration became the first in
Mexico to hire an in-house pollster (interviews). Some of the resulting datasets are now available from
the Centro de Investigación y Docencia Económicas in Mexico City.
This paragraph is based on interviews with Mexican officials, as well as Salinas de Gortari (2000).
POLLARA: Canada Perspectives, Quarter 3, 1993. N=1100.
Hart (2002: 75) notes that until the Canada-U.S. free trade agreement of the 1980s, the “basic
orientation” of Canadian trade policy included a heavy dose of protection.